Daily Papers

Vision language models (VLMs) have achieved strong performance on visual document understanding benchmarks such as DocVQA, ChartQA, and MMLongBench-Doc. However, real-world documents combine multiple factors such as length, layout complexity, modality, and question difficulty, which makes it difficult to attribute model failures to specific causes. We introduce SynthDocBench, a fully synthetic benchmark for long-context visual document understanding that systematically controls factors including document length, layout structure, modality composition, and question type. The benchmark is constructed using a combinatorial design, each factor is varied independently across generated documents, enabling controlled analysis of model behavior. Documents are generated end to end using an LLM pipeline across six layout archetypes, with a 40 percent random override to prevent models from exploiting spurious correlations. Additionally, SynthDocBench spans long-context documents with substantially greater length and structural diversity than existing benchmarks. Evaluating seven frontier VLMs, we uncover three failure modes that existing benchmarks cannot surface: sharp degradation with document length, a systematic positional sensitivity in which the middle third of a document is hardest for five of six models and five of six models show a negative Early-to-Late trend (steepest decline: 8.3 percentage points), and breakdown of chart comprehension in long-document settings. These results suggest that current models may be overfitting to benchmark artifacts rather than achieving robust long-context visual document understanding.

Real repository issues routinely include visual evidence such as screenshots, error dialogs, rendered UI states, and logs, yet repository-level issue localization is evaluated mostly as a text-only task. Existing multimodal SE benchmarks evaluate end-to-end repair, entangling localization with patch synthesis and obscuring whether visual input helped, hurt, or was ignored. We introduce MM-IssueLoc, a controlled benchmark and evaluation protocol for repository-level localization with visual evidence. MM-IssueLoc contains 652 issue-PR instances across 23 languages, with annotations for 7 image categories and 4 relevance levels. It provides file-level and function-level gold labels, paired text-only and with-image evaluation, and VCE-based diagnostics that convert images into structured textual evidence. We evaluate LLM-based and retrieval-based systems, including MM-IssueLoc-VL-Emb as a controlled multimodal retriever. Results show that existing systems remain far from reliable multimodal repository localization: the strongest agent reaches 38.96 file Acc@5 and 22.45 function Acc@10, while the strongest retriever reaches 33.86 function Acc@10. Cross-benchmark comparisons show that high localization scores on text-dominant SWE benchmarks do not transfer cleanly to multimodal issue localization. MM-IssueLoc turns visual evidence into an explicit evaluation variable, enabling future work to test whether systems improve by using visual evidence for localization, rather than by relying on text-only cues or downstream patch-generation effects.

Recent interest in multimodal large language models (MLLMs) raises a central question: can they reason over dynamic visual evidence rather than merely recognize objects or events in individual frames? This ability, which we refer to as video temporal-logical reasoning, requires models to maintain, update, and compose evidence as visual states evolve across frames. Existing video benchmarks often conflate this capability with scene complexity, static recognition, or uncontrolled temporal variation. To isolate this capability, we introduce Video-MME-Logical, a controlled benchmark organized around five temporal-logical operations: state tracking, sequential counting, temporal ordering, dynamic spatiality, and structural composition. The benchmark contains 25 fine-grained task categories generated with controlled object states, transitions, temporal dependencies, and logical compositions. It enables difficulty-controlled final-answer evaluation by varying temporal horizon and reasoning complexity, and supports intermediate-state diagnostics by verifying whether models recover the required logical reasoning trace before producing the final answer. Experiments with state-of-the-art MLLMs reveal a substantial human-model gap, especially as temporal-logical complexity increases. Supervised fine-tuning on up to 500K generated samples improves performance but remains insufficient to close the reasoning gap, positioning Video-MME-Logical as a scalable testbed for analyzing and improving temporal-logical reasoning in MLLMs.

While multi-agent systems (MAS) promise elevated intelligence through coordination of agents, current approaches to automatic MAS design under-deliver. Such shortcomings stem from two key factors: (1) methodological complexity - agent orchestration is performed using sequential, code-level execution that limits global system-level holistic reasoning and scales poorly with agent complexity - and (2) efficacy uncertainty - MAS are deployed without understanding if there are tangible benefits compared to single-agent systems (SAS). We propose MAS-Orchestra, a training-time framework that formulates MAS orchestration as a function-calling reinforcement learning problem with holistic orchestration, generating an entire MAS at once. In MAS-Orchestra, complex, goal-oriented sub-agents are abstracted as callable functions, enabling global reasoning over system structure while hiding internal execution details. To rigorously study when and why MAS are beneficial, we introduce MASBENCH, a controlled benchmark that characterizes tasks along five axes: Depth, Horizon, Breadth, Parallel, and Robustness. Our analysis reveals that MAS gains depend critically on task structure, verification protocols, and the capabilities of both orchestrator and sub-agents, rather than holding universally. Guided by these insights, MAS-Orchestra achieves consistent improvements on public benchmarks including mathematical reasoning, multi-hop QA, and search-based QA. Together, MAS-Orchestra and MASBENCH enable better training and understanding of MAS in the pursuit of multi-agent intelligence.

We introduce Grade School Math with Distracting Context (GSM-DC), a synthetic benchmark to evaluate Large Language Models' (LLMs) reasoning robustness against systematically controlled irrelevant context (IC). GSM-DC constructs symbolic reasoning graphs with precise distractor injections, enabling rigorous, reproducible evaluation. Our experiments demonstrate that LLMs are significantly sensitive to IC, affecting both reasoning path selection and arithmetic accuracy. Additionally, training models with strong distractors improves performance in both in-distribution and out-of-distribution scenarios. We further propose a stepwise tree search guided by a process reward model, which notably enhances robustness in out-of-distribution conditions.

Vision-Language Models (VLMs) have advanced multimodal understanding, yet still struggle when targets are embedded in cluttered backgrounds requiring figure-ground segregation. To address this, we introduce ChromouVQA, a large-scale, multi-task benchmark based on Ishihara-style chromatic camouflaged images. We extend classic dot plates with multiple fill geometries and vary chromatic separation, density, size, occlusion, and rotation, recording full metadata for reproducibility. The benchmark covers nine vision-question-answering tasks, including recognition, counting, comparison, and spatial reasoning. Evaluations of humans and VLMs reveal large gaps, especially under subtle chromatic contrast or disruptive geometric fills. We also propose a model-agnostic contrastive recipe aligning silhouettes with their camouflaged renderings, improving recovery of global shapes. ChromouVQA provides a compact, controlled benchmark for reproducible evaluation and extension. Code and dataset are available at https://github.com/Chromou-VQA-Benchmark/Chromou-VQA.

Multimodal large language models (MLLMs) are increasingly expected to act on visual information, yet the same scene may require different actions under different task contexts. How reliably can a model turn the same visual evidence into the action required by the current context? To answer this question, we introduce ROSE (Reference-conditioned Oddity and Symbolic Execution), a controlled benchmark that holds the visual scene fixed while varying region constraints and required symbolic outputs. Through coupled counting and coordinate-action tasks, ROSE tests whether models can infer an implicit majority reference and act on the resulting fine-grained visual evidence under changing contexts. Across nine recent MLLMs, performance drops by as much as 44.5 percentage points from counting-oriented tasks to region-conditioned action, despite 98.8\% human performance. The gap persists on paired scenes and regions for which the same model returns the correct count, while global-click and matched local controls show that coordinate grounding explains only part of the loss, revealing a distinct, model-dependent bottleneck in turning shared visual evidence into context-specific actions.

Vision-Language-Action (VLA) models achieve strong performance in robotic manipulation by leveraging pre-trained vision-language backbones. However, in downstream robotic settings, they are typically fine-tuned with limited data, leading to overfitting to specific instruction formulations and leaving robustness to paraphrased instructions underexplored. To study this gap, we introduce LIBERO-Para, a controlled benchmark that independently varies action expressions and object references for fine-grained analysis of linguistic generalization. Across seven VLA configurations (0.6B-7.5B), we observe consistent performance degradation of 22-52 pp under paraphrasing. This degradation is primarily driven by object-level lexical variation: even simple synonym substitutions cause large drops, indicating reliance on surface-level matching rather than semantic grounding. Moreover, 80-96% of failures arise from planning-level trajectory divergence rather than execution errors, showing that paraphrasing disrupts task identification. Binary success rate treats all paraphrases equally, obscuring whether models perform consistently across difficulty levels or rely on easier cases. To address this, we propose PRIDE, a metric that quantifies paraphrase difficulty using semantic and syntactic factors. Our benchmark and corresponding code are available at: https://github.com/cau-hai-lab/LIBERO-Para

Speech-based depression detection compresses features from short audio segments into one speaker-level decision, a step called temporal aggregation rarely studied on its own. Most benchmarks fix a single self-supervised encoder and a single hand-picked layer, so a reported gain may reflect the pipeline rather than the aggregation method itself. We introduce DEPOOL, a controlled benchmark that compares six aggregation architectures with six frozen speech backbones on an English and a Mandarin depression corpus, where each configuration learns which backbone layers matter rather than fixing one by hand. Across the resulting 72-configuration grid, a third of configurations collapse into predicting a single class for every speaker, a failure tied to the backbone as much as to the method, and the architecture that is most stable in a single-seed run becomes unreliable when training repeats across seeds. Robustness to backbone and seed, rather than average accuracy across a single pipeline, should be a first-class benchmarking criterion for temporal aggregation in clinical speech.

Document packets, multiple documents concatenated into a single file, are common in government and administrative workflows, yet splitting them into their constituent documents is difficult, especially for low-resource languages. We introduce Khondo (Bangla for split/segment), the first benchmark for document packet splitting on Bangladeshi government forms. Unlike prior English and OCR-text-based datasets, Khondo is bilingual (Bangla--English) and vision-native; where models operate directly on page images. It spans five concatenation schemes, from sequential to fully shuffled, across 14 administrative domains, with ground-truth boundaries, domain types, and page order. Zero-shot evaluation of MLLMs shows they cluster pages into their source documents fairly well but struggle in restoring the original page order once shuffled. To isolate what drives this difficulty, we run two controlled analyses, varying the prompt instruction and then the packet language. Both primarily affect ordering rather than clustering: (a) explicit page-order instructions are necessary but insufficient, and (b) English packets are ordered more reliably than Bangla, making page arrangement the dominant challenge and language a secondary but consistent factor. Khondo establishes page-order reconstruction as a key open problem in vision-based, low-resource document understanding, and provides a controlled benchmark for measuring progress toward solving it. Our dataset and code is available at https://huggingface.co/datasets/Mausul/khondo

W4A4 quantization promises full utilization of INT4 Tensor Cores, yet group dequantization overhead on CUDA Cores has driven existing systems to mixed-precision fallbacks. We present the first systematic study of how intra-SM compute balance governs this bottleneck. Through controlled benchmarks across four GPUs from Ampere and Ada architectures, we identify the Tensor Cores to CUDA Cores throughput ratio (ρ) as the primary hardware indicator: the W4A4-g128 kernel yields 2.0--2.5times speedup on RTX~3090 (ρ=16) yet degrades to 0.43--0.47times on A100 (ρ=64) in compute-bond scenarios, establishing W4A4 viability as platform-dependent rather than universally infeasible. Guided by this finding, we build APEX4, which co-designs pure INT4 GEMM kernels with ρ-aware granularity adaptation to mitigate the CUDA Cores dequantization bottleneck. APEX4 achieves perplexity within 0.63 of FP16 on LLaMA-2-70B and outperforms W4Ax Atom-g128 by 4.0\%--4.4\% in zero-shot accuracy. Deployed as a drop-in replacement in unmodified vLLM, it delivers up to 1.66times end-to-end speedup on L40S (ρ=8), and 1.78times on RTX~3090 (ρ=16), 2.09times on A40 (ρ=16), while recovering A100 (ρ=64) to 1.20--1.40times via the mixed-granularity mode. Our code is available at https://github.com/APEX4-W4A4/APEX4-W4A4.

The field of image generation is currently bifurcated into autoregressive (AR) models operating on discrete tokens and diffusion models utilizing continuous latents. This divide, rooted in the distinction between VQ-VAEs and VAEs, hinders unified modeling and fair benchmarking. Finite Scalar Quantization (FSQ) offers a theoretical bridge, yet vanilla FSQ suffers from a critical flaw: its equal-interval quantization can cause activation collapse. This mismatch forces a trade-off between reconstruction fidelity and information efficiency. In this work, we resolve this dilemma by simply replacing the activation function in original FSQ with a distribution-matching mapping to enforce a uniform prior. Termed iFSQ, this simple strategy requires just one line of code yet mathematically guarantees both optimal bin utilization and reconstruction precision. Leveraging iFSQ as a controlled benchmark, we uncover two key insights: (1) The optimal equilibrium between discrete and continuous representations lies at approximately 4 bits per dimension. (2) Under identical reconstruction constraints, AR models exhibit rapid initial convergence, whereas diffusion models achieve a superior performance ceiling, suggesting that strict sequential ordering may limit the upper bounds of generation quality. Finally, we extend our analysis by adapting Representation Alignment (REPA) to AR models, yielding LlamaGen-REPA. Codes is available at https://github.com/Tencent-Hunyuan/iFSQ

Segmentation is a fundamental vision task underlying numerous downstream applications. Recent promptable segmentation models, such as Segment Anything Model 3 (SAM3), extend segmentation from category-agnostic mask prediction to concept-guided localization conditioned on high-level textual prompts. However, existing benchmarks primarily evaluate mask accuracy or object presence, leaving unclear whether these models faithfully ground the queried concept or instead rely on visually salient but semantically misleading cues. We introduce CAFE: Counterfactual Attribute Factuality Evaluation, a novel benchmark for evaluating concept-faithful segmentation in promptable segmentation models. Our CAFE is built on attribute-level counterfactual manipulation: the target region and ground-truth mask are preserved, while attributes such as surface appearance, context, or material composition are modified to introduce misleading semantic cues. The benchmark contains 2,146 paired test samples, each consisting of a target image, a ground-truth mask, a positive prompt, and a misleading negative prompt. These samples cover three counterfactual categories: Superficial Mimicry (SM), Context Conflict (CC), and Ontological Conflict (OC). We evaluate various model types and sizes on our CAFE. Experiments reveal a systematic gap between localization quality and concept discrimination: models often generate accurate masks even for misleading prompts, suggesting that strong mask prediction does not necessarily imply faithful semantic grounding. Our CAFE provides a controlled benchmark for diagnosing whether promptable segmentation models perform concept-faithful grounding rather than shortcut-driven mask retrieval.

We present a novel and practical deep fully convolutional neural network architecture for semantic pixel-wise segmentation termed SegNet. This core trainable segmentation engine consists of an encoder network, a corresponding decoder network followed by a pixel-wise classification layer. The architecture of the encoder network is topologically identical to the 13 convolutional layers in the VGG16 network. The role of the decoder network is to map the low resolution encoder feature maps to full input resolution feature maps for pixel-wise classification. The novelty of SegNet lies is in the manner in which the decoder upsamples its lower resolution input feature map(s). Specifically, the decoder uses pooling indices computed in the max-pooling step of the corresponding encoder to perform non-linear upsampling. This eliminates the need for learning to upsample. The upsampled maps are sparse and are then convolved with trainable filters to produce dense feature maps. We compare our proposed architecture with the widely adopted FCN and also with the well known DeepLab-LargeFOV, DeconvNet architectures. This comparison reveals the memory versus accuracy trade-off involved in achieving good segmentation performance. SegNet was primarily motivated by scene understanding applications. Hence, it is designed to be efficient both in terms of memory and computational time during inference. It is also significantly smaller in the number of trainable parameters than other competing architectures. We also performed a controlled benchmark of SegNet and other architectures on both road scenes and SUN RGB-D indoor scene segmentation tasks. We show that SegNet provides good performance with competitive inference time and more efficient inference memory-wise as compared to other architectures. We also provide a Caffe implementation of SegNet and a web demo at http://mi.eng.cam.ac.uk/projects/segnet/.

AI research often requires decisions before future evidence exists: which bottleneck to attack, which direction to pursue, or where a project should be positioned. We introduce ForeSci, a temporally controlled benchmark for evaluating whether LLM agents can make such forward-looking research judgements from historical evidence. ForeSci contains 500 tasks across four fast-moving AI domains and four decision families. Each task is paired with a cutoff-aligned offline knowledge base; post-cutoff papers are hidden during generation and used only for validation. To avoid random future-event prediction, tasks are derived from pre-cutoff taxonomy branches and evidence signals, and answer-generation backbones are selected to precede the task cutoffs. We evaluate native LLMs, Hybrid RAG, and three research-agent adaptations across four backbones. Results show that explicit evidence organization improves traceability and factual support, but gains depend strongly on the decision family. Diagnostics reveal a recurring evidence-decision decoupling: agents may cite relevant evidence while forecasting the wrong research object. ForeSci turns forward-looking AI research judgement into a controlled benchmark for evaluating research agents as decision-making systems.

We present TornadoNet, a comprehensive benchmark for automated street-level building damage assessment evaluating how modern real-time object detection architectures and ordinal-aware supervision strategies perform under realistic post-disaster conditions. TornadoNet provides the first controlled benchmark demonstrating how architectural design and loss formulation jointly influence multi-level damage detection from street-view imagery, delivering methodological insights and deployable tools for disaster response. Using 3,333 high-resolution geotagged images and 8,890 annotated building instances from the 2021 Midwest tornado outbreak, we systematically compare CNN-based detectors from the YOLO family against transformer-based models (RT-DETR) for multi-level damage detection. Models are trained under standardized protocols using a five-level damage classification framework based on IN-CORE damage states, validated through expert cross-annotation. Baseline experiments reveal complementary architectural strengths. CNN-based YOLO models achieve highest detection accuracy and throughput, with larger variants reaching 46.05% mAP@0.5 at 66-276 FPS on A100 GPUs. Transformer-based RT-DETR models exhibit stronger ordinal consistency, achieving 88.13% Ordinal Top-1 Accuracy and MAOE of 0.65, indicating more reliable severity grading despite lower baseline mAP. To align supervision with the ordered nature of damage severity, we introduce soft ordinal classification targets and evaluate explicit ordinal-distance penalties. RT-DETR trained with calibrated ordinal supervision achieves 44.70% mAP@0.5, a 4.8 percentage-point improvement, with gains in ordinal metrics (91.15% Ordinal Top-1 Accuracy, MAOE = 0.56). These findings establish that ordinal-aware supervision improves damage severity estimation when aligned with detector architecture. Model & Data: https://github.com/crumeike/TornadoNet

World models for embodied AI must be physically viable: constructed to answer intervention queries by representing the physical structure governing action outcomes, rather than merely predicting future observations. Existing observation-predictive world models can produce visually plausible but physically wrong rollouts. This failure is structural; distinct physical systems can look identical yet diverge under intervention. We expose this problem with controlled benchmarks that fix the visible scene while varying latent physics. We show that such models may recommend infeasible actions, mispredict interaction outcomes, or certify unsafe behavior. We argue that embodied AI requires world models that identify the simplest physical abstraction sufficient to answer an intervention query. Such a model comprises modular components, including environment representation, latent state and parameter estimation, action specification, interventional dynamics, and query-level response. An autonomous orchestrator should identify the relevant abstraction and compose compatible learned and structured components per query. When closed-form physics is unavailable, uncertain, or costly, the transition model may be analytic, simulated, learned, or hybrid, but it must preserve the structure that determines interventional outcomes. This decomposition makes the model interpretable, its components verifiable, and its outputs auditable against the query. It also provides a design principle for new world models and a feasibility test for existing ones: the right abstraction is not the most detailed model of the world, but the simplest model that preserves the distinctions relevant to the query. We demonstrate this approach on queries that existing systems fail to answer correctly, and outline how an orchestrator can dynamically assemble and adapt physically viable models for planning, control, and verification.

Multimodal large language models (MLLMs) are increasingly deployed in personally and societally consequential settings, yet the visual cues that shape how these models judge people remain poorly understood. Prior work often compares different (groups of) individuals, making it difficult to separate appearance effects from identity differences. We introduce StylisticBias, a controlled benchmark for evaluating attribute-level social bias in MLLMs. We generate 500 photorealistic base faces and create about 50 single-attribute variations per face, producing about 25K images. This design keeps identity fixed and changes one visual attribute at a time. It lets us measure how specific cues shift model judgments. We evaluate six MLLMs across 25 binary social judgment scenarios. We find that age and body type dominate identity-level effects, while fashion style and other visual cues drive the largest attribute-level shifts. We further find that about 15 attributes account for nearly 80\% of the total variation, showing that bias is concentrated in a small set of visual cues. Sensitivity is strongest in judgments that are semantically aligned with appearance, especially socioeconomic and style-related judgments. We release StylisticBias as a benchmark for fine-grained bias evaluation in multimodal models. Code and dataset: https://github.com/timo-cavelius/StylisticBias and https://hf.co/datasets/shaghayegh/stylistic-bias-dataset.

Large language models often default to step-by-step computation even when efficient numerical shortcuts are available. This raises a basic question: do they exhibit number sense in a human-like behavioral sense, i.e., the ability to recognize numerical structure, apply shortcuts when appropriate, and avoid them when they are not? We introduce SenseMath, a controlled benchmark for evaluating structure-sensitive numerical reasoning in LLMs. SenseMath contains 4,800 items spanning eight shortcut categories and four digit scales, with matched strong-shortcut, weak-shortcut, and control variants. It supports three evaluation settings of increasing cognitive demand: Shortcut Use (whether models can apply shortcuts on shortcut-amenable problems); Applicability Judgment (whether they can recognize when a shortcut is appropriate or misleading); and Problem Generation (whether they can generate new problem items that correctly admit a given type of shortcut). Our evaluation across five LLMs, ranging from GPT-4o-mini to Llama-3.1-8B, shows a consistent pattern: when explicitly prompted, models readily adopt shortcut strategies and achieve substantial accuracy gains on shortcut-amenable items (up to 15%), yet under standard chain-of-thought prompting they spontaneously employ such strategies in fewer than 40% of cases, even when they demonstrably possess the requisite capability. Moreover, this competence is confined to the Use level; models systematically over-generalise shortcuts to problems where they do not apply, and fail to generate valid shortcut-bearing problems from scratch. Together, these results suggest that current LLMs exhibit procedural shortcut fluency without the structural understanding of when and why shortcuts work that underlies human number sense.

Large language model (LLM)-based coding agents achieve impressive results on controlled benchmarks yet routinely produce pull requests that real maintainers reject. The root cause is not functional incorrectness but a lack of organicity: generated code ignores project-specific conventions, duplicates functionality already provided by internal APIs, and violates implicit architectural constraints accumulated over years of development. Simply exposing an agent to the latest repository snapshot is not enough: the snapshot reveals the final state of the codebase, but not the repository-specific change patterns by which that state was reached. We introduce Learning to Commit, a framework that closes this gap through Online Repository Memory. Given a repository with a strict chronological split, the agent performs supervised contrastive reflection on earlier commits: it blindly attempts to resolve each historical issue, compares its prediction against the oracle diff, and distils the gap into a continuously growing set of skills-reusable patterns capturing coding style, internal API usage, and architectural invariants. When a new PR description arrives, the agent conditions its generation on these accumulated skills, producing changes grounded in the project's own evolution rather than generic pretraining priors. Evaluation is conducted on genuinely future, merged pull requests that could not have been seen during the skill-building phase, and spans multiple dimensions including functional correctness, code-style consistency, internal API reuse rate, and modified-region plausibility. Experiments on an expert-maintained repository with rich commit history show that Online Repository Memory effectively improves organicity scores on held-out future tasks.

Vision-Language Models (VLMs) excel on many multimodal reasoning benchmarks, but these evaluations often do not require an exhaustive readout of the image and can therefore obscure failures in faithfully capturing all visual details. We introduce Grid2Matrix (G2M), a controlled benchmark in which a model is shown a color grid and a color-to-number mapping, and must output the corresponding matrix. By varying grid size and the number of colors, G2M provides a simple way to increase visual complexity while minimizing semantic confounds. We find that VLMs exhibit a sharp early collapse in zero-shot end-to-end evaluation, failing on surprisingly small grids rather than degrading gradually as the task becomes denser. We probe the visual encoders of VLMs from two representative families and find that they preserve substantially more of the grid information than the corresponding end-to-end outputs. This suggests that the failure is not explained by visual encoding alone, but also reflects a gap between what remains recoverable from visual features and what is ultimately expressed in language. We term this gap Digital Agnosia. Further analyses show that these errors are highly structured and depend strongly on how grid cells overlap with visual patch boundaries. We also find that common strategies such as model scaling and multimodal alignment do not fully eliminate this failure mode. We expect G2M to serve as a useful testbed for understanding where and how VLMs lose fine visual details, and for evaluating tasks where missing even small visual details can matter, such as tables, charts, forms, and GUIs.

Deploying language-model agents in production often requires substantial compute and human effort to tune prompts, parsers, validators, and other components of the agent pipeline. Self-evolution offers a promising alternative, but most existing frameworks assume access to frontier models that can reliably diagnose failures, propose revisions, and judge their own updates. We study whether frozen small language models (SLMs) can serve as effective self-evolving agents under resource constraints. We propose PACE (Prompt And Control Logic Evolution), a two-timescale framework that coordinates low-risk prompt refinement with higher-risk control-logic updates. PACE evolves prompts under fixed control logic until prompt-level gains saturate, then considers constrained control-logic updates that are accepted through held-out validation. Across three frozen SLM backbones ranging from 4B to 14B parameters and four controlled benchmarks, PACE achieves the best performance on all 12 backbone--benchmark combinations, improving over vanilla SLM agents by up to +9.2% relative improvement and over the stronger single-mode evolution baseline by up to +5.4% relative improvement. A tau-bench case study further shows that PACE improves multi-turn tool-use success over vanilla and prompt-only evolution. These results suggest that reliable SLM agent self-evolution is possible without updating model weights or relying on frontier-model teachers, and that the key benefit is not any single final solver pattern but autonomous, validated discovery of task-appropriate inference strategies.

We study the problem of training self-supervised foundation models for functional MRI. Our main contributions are: (1) we introduce a new model family (CortexMAE) trained using the masked autoencoder framework on 2.1K hours of open fMRI data, and (2) we release the first open evaluation suite (Brainmarks) for fMRI foundation models. Our core innovation is simple: we adapt the Vision Transformer to fMRI by first converting each 3D fMRI volume to a 2D map using a cortical flat map projection. We directly compare flat maps to both parcellation and volume-based representations. While each has its advantages, flat maps generally perform best. We perform the first systematic scaling analysis for fMRI and observe strict power law scaling, albeit with limits. Finally, we use Brainmarks to do controlled benchmark comparisons. On subject-level trait prediction, we report a challenging null result: no single model achieves clear state-of-the-art performance. Moreover, all models struggle to outperform a simple functional connectivity baseline. On cognitive state decoding, we observe more robust performance, and in this setting our CortexMAE family outperforms prior models by a large margin. Code, models, and datasets are available at https://github.com/MedARC-AI/CortexMAE and https://github.com/MedARC-AI/Brainmarks.

Deep research agents are increasingly evaluated on their ability to search for evidence, reason over retrieved sources, and produce grounded answers. Existing browsing benchmarks, however, largely assume that the user's query and the supporting evidence are written in the same language, leaving open whether agentic search systems can operate when relevant evidence appears in another language. We introduce XBCP (Cross-lingual BrowseComp-Plus), a controlled benchmark that preserves the English question-and-answer space of BrowseComp-Plus but varies the languages of the supporting documents. XBCP instantiates two complementary settings: in the cross-lingual setting, each query is paired with evidence in a single assigned language. In the multilingual setting, the full evidence corpus is distributed equally and randomly across 12 languages spanning high-resource and low-resource regimes. We evaluate four deep research agents using sparse and dense multilingual retrievers, measuring answer accuracy, evidence recall, search behavior, calibration, citation fidelity, and oracle retrieval. Results reveal substantial degradation when evidence is translated. Even strong, dense retrievers lose evidence recall, and agents become less calibrated and cite evidence less reliably. Notably, accuracy remains lower even when all gold evidence is supplied directly. These findings suggest that cross-lingual deep research exposes both retrieval failures and an independent, agent-side difficulty in integrating language-mismatched evidence.

Frontier language models sometimes recognize that they are being evaluated and adjust their behavior, undermining validity of benchmark results. Yet the field studies it without a shared foundation, conflating properties of the evaluation with properties of the model, and detection with behavioral response. We ground evaluation awareness in social psychology, decomposing it into an environment component (how recognizable the task is) and a model component that separates recognition from propensity to act on it. We operationalize the environment component through eight categorized trigger factors, such as placeholder entities and grading-style output formats, and study recognition and behavior through chain-of-thought monitoring. Across nine frontier models and four benchmarks, recognition rates depend on the specific pairing of model and benchmark rather than on either in isolation. Recognition rarely leads to behavioral change, and when it does, the direction depends on the type of evaluation perceived. Models are also more sensitive to safety than capability evaluations, placing safety benchmark validity at greater risk. To study which factors each model is sensitive to and how they interact, we propose EvalAwareBench, a factor-controlled benchmark of 100 paired safety-capability tasks where each of the eight factors can be independently toggled, varying evaluative signals while holding the underlying request fixed. Through EvalAwareBench, we find that no single factor uniformly affects all models, but stacking factors progressively raises evaluation awareness across all of them. Our framework and EvalAwareBench provide the tools to measure, attribute, and mitigate evaluation awareness, pointing to behavioral consistency under recognition as a promising path forward.

Understanding emotion from speech requires sensitivity to both lexical and acoustic cues. However, it remains unclear whether large audio language models (LALMs) genuinely process acoustic information or rely primarily on lexical content. We present LISTEN (Lexical vs. Acoustic Speech Test for Emotion in Narratives), a controlled benchmark designed to disentangle lexical reliance from acoustic sensitivity in emotion understanding. Across evaluations of six state-of-the-art LALMs, we observe a consistent lexical dominance. Models predict "neutral" when lexical cues are neutral or absent, show limited gains under cue alignment, and fail to classify distinct emotions under cue conflict. In paralinguistic settings, performance approaches chance. These results indicate that current LALMs largely "transcribe" rather than "listen," relying heavily on lexical semantics while underutilizing acoustic cues. LISTEN offers a principled framework for assessing emotion understanding in multimodal models.

Multi-server MCP agents create an information-flow control problem: faithful tool composition can turn individually benign read/write permissions into cross-boundary credential propagation -- a structural side effect of workflow topology, not necessarily malicious model behavior. We present MCPHunt, to our knowledge the first controlled benchmark that isolates non-adversarial, verbatim credential propagation across multi-server MCP trust boundaries, with three methodological contributions: (1) canary-based taint tracking that reduces propagation detection to objective string matching; (2) an environment-controlled coverage design with risky, benign, and hard-negative conditions that validates pipeline soundness and controls for credential-format confounds; (3) CRS stratification that disentangles task-mandated propagation (faithful execution of verbatim-transfer instructions) from policy-violating propagation (credentials included despite the option to redact). Across 3,615 main-benchmark traces from 5 models spanning 147 tasks and 9 mechanism families, policy-violating propagation rates reach 11.5--41.3% across all models. This propagation is pathway-specific (25x cross-mechanism range) and concentrated in browser-mediated data flows; hard-negative controls provide evidence that production-format credentials are not necessary -- prompt-directed cross-boundary data flow is sufficient. A prompt-mitigation study across 3 models reduces policy-violating propagation by up to 97% while preserving 80.5% utility, but effectiveness varies with instruction-following capability -- suggesting that prompt-level defenses alone may not suffice. Code, traces, and labeling pipeline are released under MIT and CC BY 4.0.

Continuous perception, the ability to integrate visual observations over time in a continuous stream fashion, is essential for robust real-world understanding, yet remains largely untested in current multimodal models. We introduce CP-Bench, a minimal and fully controlled benchmark designed to isolate this capability using an extremely simple task: counting identical cubes in a synthetic scene while the camera moves and only reveals subsets of objects at any moment. Despite the simplicity of the setting, we find that state-of-the-art open-source and commercial models, including Qwen-3-VL, InternVL3, GPT-5, and Gemini-3-Pro, fail dramatically. A static-camera control variant confirms that the failure arises not from object recognition but from an inability to accumulate evidence across time. Further experiments show that neither higher sampling FPS, perception- or spatial-enhanced models, nor finetuning with additional videos leads to meaningful cross-temporal generalization. Our results reveal a fundamental limitation in modern multimodal architectures and training paradigms. CP-Bench provides a simple yet powerful diagnostic tool and establishes a clean testbed for developing models capable of genuine time-consistent visual reasoning.

Matrix-based optimizers have attracted growing interest for improving LLM training efficiency, with significant progress centered on orthogonalization/whitening based methods. While yielding substantial performance gains, a fundamental question arises: can we develop new paradigms beyond orthogonalization, pushing the efficiency frontier further? We present Adaptively Rotated Optimization (ARO, a new matrix optimization framework that treats gradient rotation as a first class design principle. ARO accelerates LLM training by performing normed steepest descent in a rotated coordinate system, where the rotation is determined by a novel norm-informed policy. This perspective yields update rules that go beyond existing orthogonalization and whitening optimizers, improving sample efficiency in practice. To make comparisons reliable, we propose a rigorously controlled benchmarking protocol that reduces confounding and bias. Under this protocol, ARO consistently outperforms AdamW (by 1.3 sim1.35times) and orthogonalization methods (by 1.1sim1.15times) in LLM pretraining at up to 8B activated parameters, and up to 8times overtrain budget, without evidence of diminishing returns. Finally, we discuss how ARO can be reformulated as a symmetry-aware optimizer grounded in rotational symmetries of residual streams, motivating advanced designs that enable computationally efficient exploitation of cross-layer/cross module couplings.

Memory is becoming a core component of long-horizon AI agents, allowing agents to reuse past experience when operating web browsers, software tools, and other interactive environments. Existing work mostly treats memory as a supply problem, asking what experience to write, how to store it, and which entry to retrieve for the next task. Yet we still lack a clear account of how models consume retrieved memory across a multi-step action trajectory. This consumption process matters because it determines not only what memories should be retrieved, but also what models and control policies are needed to use them safely. To diagnose this process, we propose Entry--Propagation--Recovery (E-P-R), a trajectory-level framework that asks where memory first changes an action, whether that change carries forward, and whether the agent can recover after leaving a correct path. We instantiate E-P-R on WebArena and on MemTrapBench, a controlled benchmark we build to isolate these phases. We find that the main failure often begins at entry: agents adopt conflicting memory at the first exposed decision point even when it is task-wrong. Repeated exposure then amplifies this early error, while recovery after divergence is weak. Together, these effects create a compliance trap: across models, conflicting memory induces similar compliance rates, but once agents comply, their success rates collapse to a low floor. Stronger agents therefore suffer larger absolute damage because each compliance event erases more baseline capability. These results suggest that memory-augmented agents should be evaluated not only by retrieval quality or final success rate, but by how they consume memory throughout the trajectory.

Multimodal agents are increasingly expected to operate interfaces on behalf of users, raising a central deployment question: can they truly substitute for humans in workflows that services deliberately protect against automation? CAPTCHA verification makes this question concrete. It is not merely a visual puzzle, but a human-verification boundary placed before account creation, content access, form submission, and other protected actions. We introduce Humanity's Last Line of Verification (HLL), a controlled benchmark that uses interactive CAPTCHA verification to evaluate whether agents can cross this boundary through grounded, human-like interaction rather than recognition alone. HLL covers diverse CAPTCHA interactions and exposes agents to controlled realism stressors, including cluttered webpages, harder task variants, and trace-conditioned validation of the solving process. We evaluate eight frontier multimodal agents in a closed-loop GUI environment. The results show that current agents remain brittle at this human-substitution boundary: performance varies sharply across verification types, degrades under realistic interface conditions, and drops further when correct answers must be supported by valid action traces. By exposing gaps in localization, action calibration, state tracking, and process consistency, HLL provides a concrete testbed for measuring how close multimodal agents are to acting as human substitutes in protected real-world workflows. Our code is available at https://github.com/XinhaoS0101/HLL

Anonymization is often assumed to protect privacy once explicit identifiers are removed, because re-identification has historically required specialized expertise, tailored algorithms, and manual corroboration. We show that LLM-based agents weaken this barrier: by combining scattered, individually non-identifying cues with public evidence, they reconstruct real-world identities, sometimes even during benign tasks. We evaluate this risk across three settings -- classical linkage incidents, a controlled benchmark (InferLink) that varies fingerprint type, task framing, and attacker knowledge, and open-ended human--AI interaction traces. In the sparsest regime of the Netflix Prize deanonymization setting, agents reconstruct 79.2\% of identities, against 56.0\% for a classical matching baseline; on InferLink, they link individuals even without an explicit re-identification request, and more often once one is given. In redacted human--AI interaction traces, agents further resolve anonymized profiles to specific individuals by corroborating contextual cues with public evidence. These findings suggest that privacy evaluations for agentic systems should measure not only what information is accessed or disclosed, but also what identities can be inferred.

As generative Artificial Intelligence (AI) advances, the realism of AI generated imagery has reached a threshold capable of deceiving even vigilant human observers. Yet, while current AI-generated Image Detection (AID) approaches perform exceptionally well on controlled benchmark datasets, they struggle significantly with real-world cases. To study this behavior we introduce the ITW-SM dataset, a curated collection of real and AI-generated images originating from major social media platforms. We employ it to analyze the effects of key design choices typically considered when building a detector, involving its architecture, pre-trained latent spaces, training data as well as pre-processing approaches. We indicate that naively scaling the pre-training stage or opting for more training data does not always lead to better detection performance. Instead, our work reveals that it is crucial to optimize each design choice to enable the processing pipeline to propagate and effectively analyze both low-level traces as well as high-level image semantics. Building on our findings, we achieve a substantial average improvement of 26.87% in AUC across multiple state-of-the-art detection approaches and under real-world conditions, providing a roadmap for developing more resilient detectors. Our assets are available on https://mever-team.github.io/itw-sm.

Large vision-language models (VLMs) often rely on familiar semantic priors, but existing evaluations do not cleanly separate perception failures from rule-mapping failures. We study this behavior as semantic fixation: preserving a default interpretation even when the prompt specifies an alternative, equally valid mapping. To isolate this effect, we introduce VLM-Fix, a controlled benchmark over four abstract strategy games that evaluates identical terminal board states under paired standard and inverse rule formulations. Across 14 open and closed VLMs, accuracy consistently favors standard rules, revealing a robust semantic-fixation gap. Prompt interventions support this mechanism: neutral alias prompts substantially narrow the inverse-rule gap, while semantically loaded aliases reopen it. Post-training is strongly rule-aligned: training on one rule improves same-rule transfer but hurts opposite-rule transfer, while joint-rule training improves broader transfer. To test external validity beyond synthetic games, we evaluate analogous defamiliarization interventions on VLMBias and observe the same qualitative pattern. Finally, late-layer activation steering partially recovers degraded performance, indicating that semantic-fixation errors are at least partly editable in late representations. Project page, code, and dataset available at https://maveryn.github.io/vlm-fix/.

We present Urban-ImageNet, a large-scale multi-modal dataset and evaluation benchmark for urban space perception from user-generated social media imagery. The corpus contains over 2 Million public social media images and paired textual posts collected from Weibo across 61 urban sites in 24 Chinese cities across 2019-2025, with controlled benchmark subsets at 1K, 10K, and 100K scale and a full 2M corpus for large-scale training and evaluation. Urban-ImageNet is organized by HUSIC, a Hierarchical Urban Space Image Classification framework that defines a 10-class taxonomy grounded in urban theory. The taxonomy is designed to distinguish activated and non-activated public spaces, exterior and interior urban environments, accommodation spaces, consumption content, portraits, and non-spatial social-media content. Rather than treating urban imagery as generic scene data, Urban-ImageNet evaluates whether machine perception models can capture spatial, social, and functional distinctions that are central to urban studies. The benchmark supports three tasks within one standardized library: (T1) urban scene semantic classification, (T2) cross-modal image-text retrieval, and (T3) instance segmentation. Our experiments evaluate representative vision, vision-language, and segmentation models, revealing strong performance on supervised scene classification but more challenging behavior in cross-modal retrieval and instance-level urban object segmentation. A multi-scale study further examines how model performance changes as balanced training data increases from 1K, 10K to 100K images. Urban-ImageNet provides a unified, theory-grounded, multi-city benchmark for evaluating how AI systems perceive and interpret contemporary urban spaces across modalities, scales, and task formulations. Dataset and benchmark are available at: huggingface.co/datasets/Yiwei-Ou/Urban-ImageNet and github.com/yiasun/dataset-2.

A significant gap exists between theory and practice in deep learning. Generalization and approximation error bounds are often derived for simplified models or are too loose to be informative. Many rely on the manifold hypothesis and on geometric regularity such as intrinsic dimension, curvature, and reach. Progress requires insight into data-manifold geometry and suitable benchmarks, yet existing options are polarized: analytic manifolds with known geometry but limited applicability, or real-world datasets where geometry is only coarsely estimable. We introduce a benchmarking framework for studying data geometry. We repurpose and extend dSprites and COIL-20 with additional transformation dimensions and dense, axis-aligned sampling, and pair them with finite-difference estimators that recover curvature, reach, and volume at near-ground-truth accuracy in a regime where general-purpose estimators are unreliable or difficult to deploy. The framework is intended as a controlled testbed, useful as a calibration environment for geometric estimators and a sandbox for probing theoretical assumptions. To illustrate its use, we present two application studies, namely assessing the scaling behavior of the bounds of Genovese et al. and Fefferman et al., and tracking the layer-wise geometry of a β-VAE, highlighting the behavior of current bounds and the value of controlled benchmarks for guiding and validating future theory. A reference implementation is available at https://github.com/koulakis/manifold-microscope.

The rapid adoption of LLMs has increased the need for reliable AI text detection, yet existing detectors often fail outside controlled benchmarks. We systematically evaluate 2 dominant paradigms (training-free and supervised) and show that both are brittle under distribution shift, unseen generators, and simple stylistic perturbations. To address these limitations, we propose a supervised contrastive learning (SCL) framework that learns discriminative style embeddings. Experiments show that while supervised detectors excel in-domain, they degrade sharply out-of-domain, and training-free methods remain highly sensitive to proxy choice. Overall, our results expose fundamental challenges in building domain-agnostic detectors. Our code is available at: https://github.com/HARSHITJAIS14/DetectAI

The rapid expansion of capabilities in Large Language Model (LLM) agents has exposed a critical architectural bottleneck: when agents are given access to a flat, monolithic registry of tools, the model must evaluate hundreds or thousands of options simultaneously. This leads to decision-space explosion, context window saturation, and degraded routing accuracy. To address these limitations, this paper presents a hierarchical, skill-based architecture for agentic orchestration. Capabilities are organized as a rooted tree where internal nodes make routing decisions and leaf nodes execute deterministic tasks. The runtime enforces a single-step execution loop governed by a Last-In-First-Out (LIFO) stack, giving the agent a form of memory akin to a Pushdown Automaton, therefore enabling it to track nested execution contexts and resume deterministically from any depth. Capability discovery follows a manifest-driven, lazy-loading protocol: only the immediate children of the active node are loaded, so memory and prompt costs scale with the explored path rather than the global registry. By replacing global memory with localized stack frames, the architecture prevents outputs from one execution branch from leaking into another, establishing the isolation guarantees required for deployment in regulated enterprise environments. We also discuss UPI Help, an AI-powered digital payments support product, as a motivating production deployment context. We provide a mathematical formalization of the orchestration state, detailed algorithmic analysis of the execution loop, and controlled benchmarks comparing flat and hierarchical routing under increasing tool catalogs, multi-step workflow pressure, and visible schema-token exposure per LLM call.

Egocentric video action recognition under domain shifts remains challenging due to large intra-class spatio-temporal variability, long-tailed feature distributions, and strong correlations between actions and environments. Existing benchmarks for egocentric domain generalization often conflate covariate shifts with concept shifts, making it difficult to reliably evaluate a model's ability to generalize across input distributions. To address this limitation, we introduce Ego4OOD, a domain generalization benchmark derived from Ego4D that emphasizes measurable covariate diversity while reducing concept shift through semantically coherent, moment-level action categories. Ego4OOD spans eight geographically distinct domains and is accompanied by a clustering-based covariate shift metric that provides a quantitative proxy for domain difficulty. We further leverage a one-vs-all binary training objective that decomposes multi-class action recognition into independent binary classification tasks. This formulation is particularly well-suited for covariate shift by reducing interference between visually similar classes under feature distribution shift. Using this formulation, we show that a lightweight two-layer fully connected network achieves performance competitive with state-of-the-art egocentric domain generalization methods on both Argo1M and Ego4OOD, despite using fewer parameters and no additional modalities. Our empirical analysis demonstrates a clear relationship between measured covariate shift and recognition performance, highlighting the importance of controlled benchmarks and quantitative domain characterization for studying out-of-distribution generalization in egocentric video.

In egocentric action recognition a single population model is typically trained and subsequently embodied on a head-mounted device, such as an augmented reality headset. While this model remains static for new users and environments, we introduce an adaptive paradigm of two phases, where after pretraining a population model, the model adapts on-device and online to the user's experience. This setting is highly challenging due to the change from population to user domain and the distribution shifts in the user's data stream. Coping with the latter in-stream distribution shifts is the focus of continual learning, where progress has been rooted in controlled benchmarks but challenges faced in real-world applications often remain unaddressed. We introduce EgoAdapt, a benchmark for real-world egocentric action recognition that facilitates our two-phased adaptive paradigm, and real-world challenges naturally occur in the egocentric video streams from Ego4d, such as long-tailed action distributions and large-scale classification over 2740 actions. We introduce an evaluation framework that directly exploits the user's data stream with new metrics to measure the adaptation gain over the population model, online generalization, and hindsight performance. In contrast to single-stream evaluation in existing works, our framework proposes a meta-evaluation that aggregates the results from 50 independent user streams. We provide an extensive empirical study for finetuning and experience replay.

Medication reconciliation at clinical handoffs is a high-stakes, error-prone process. Large language models are increasingly proposed to assist with this task using FHIR-structured patient records, but a fundamental and largely unstudied variable is how the FHIR data is serialised before being passed to the model. We present the first systematic comparison of four FHIR serialisation strategies (Raw JSON, Markdown Table, Clinical Narrative, and Chronological Timeline) across five open-weight models (Phi-3.5-mini, Mistral-7B, BioMistral-7B, Llama-3.1-8B, Llama-3.3-70B) on a controlled benchmark of 200 synthetic patients, totalling 4,000 inference runs. We find that serialisation strategy has a large, statistically significant effect on performance for models up to 8B parameters: Clinical Narrative outperforms Raw JSON by up to 19 F1 points for Mistral-7B (r = 0.617, p < 10^{-10}). This advantage reverses at 70B, where Raw JSON achieves the best mean F1 of 0.9956. In all 20 model and strategy combinations, mean precision exceeds mean recall: omission is the dominant failure mode, with models more often missing an active medication than fabricating one, which changes how clinical safety auditing priorities should be set. Smaller models plateau at roughly 7-10 concurrent active medications, leaving polypharmacy patients, the patients most at risk from reconciliation errors, systematically underserved. BioMistral-7B, a domain-pretrained model without instruction tuning, produces zero usable output in all conditions, showing that domain pretraining alone is not sufficient for structured extraction. These results offer practical, evidence-based format recommendations for clinical LLM deployment: Clinical Narrative for models up to 8B, Raw JSON for 70B and above. The complete pipeline is reproducible on open-source tools running on an AWS g6e.xlarge instance (NVIDIA L40S, 48 GB VRAM).

Dynamic-scene reconstruction is almost always evaluated inside the observed time window, yet deployment settings such as AR overlays, robot interaction, and anticipatory planning need the future surface: the geometry at times beyond those captured. No standard benchmark measures this. We introduce FutureSurf, a controlled diagnostic benchmark and dataset for future-time surface reconstruction that trades scene diversity for exact future ground truth and falsification controls. A method trains on the observed first 75% of a sequence; we score its extracted per-frame surface on the held-out future by Chamfer distance, reporting absolute future CD as the primary score and the future/observed gap as a diagnostic. The dataset contains eight analytically defined controlled motions, including three falsification controls, with exact per-frame ground-truth meshes. We also provide a ground-truth-side recoverability oracle. The release includes split files, scoring code, a benchmark card, and Croissant metadata. On the controlled motions, the DG-Mesh backbone leaves a 2.7-4.1times gap even for futures predictable in principle (four of five recoverable from observed motion by a fixed rule), while the falsification controls behave as designed (the surface-invariant motion shows no gap). Beyond the contributed dataset, the gap persists across six animated DG-Mesh asset scenes and a second backbone, Deformable-3DGS (2.0-6.6times; both share a deformation-MLP temporal model). The benchmark also shows that future rendering quality and future-surface accuracy are statistically decoupled, so the novel-view-synthesis metrics the field reports do not track future geometry. The future error is structured, concentrating where the surface moves. The dataset, evaluation toolkit, and scoring code are available on Hugging Face and GitHub (https://github.com/Ricky-S/futuresurf).

As the field of Graph Neural Networks (GNN) continues to grow, it experiences a corresponding increase in the need for large, real-world datasets to train and test new GNN models on challenging, realistic problems. Unfortunately, such graph datasets are often generated from online, highly privacy-restricted ecosystems, which makes research and development on these datasets hard, if not impossible. This greatly reduces the amount of benchmark graphs available to researchers, causing the field to rely only on a handful of publicly-available datasets. To address this problem, we introduce a novel graph generative model, Computation Graph Transformer (CGT) that learns and reproduces the distribution of real-world graphs in a privacy-controlled way. More specifically, CGT (1) generates effective benchmark graphs on which GNNs show similar task performance as on the source graphs, (2) scales to process large-scale graphs, (3) incorporates off-the-shelf privacy modules to guarantee end-user privacy of the generated graph. Extensive experiments across a vast body of graph generative models show that only our model can successfully generate privacy-controlled, synthetic substitutes of large-scale real-world graphs that can be effectively used to benchmark GNN models.

The KV-cache is the right memory for datacenters but the wrong memory for robots. Datacenter inference batches many short requests and resets them, amortizing an attention cache across a crowd. Embodied agents instead run one long, non-resetting episode on bandwidth-limited edge hardware, where high-bandwidth memory and flash are scarce, flash has finite write endurance, and memory writes rather than compute can become the binding constraint. AURA-Mem (Action-Utility Recurrent Adaptive Memory) targets this regime. It wraps a frozen vision-language-action backbone with a constant-size recurrent memory and a learned gate that writes only when the current observation would change the next action: memory that knows when to stay silent. Unlike reconstruction-based memory, the gate is trained directly against a closed-loop action-error signal. Its inference state is fixed at 4,224 bytes regardless of horizon, while a KV-cache grows to 6,061 times larger at 100,000 steps. On a controlled synthetic benchmark, AURA-Mem matches the best O(1) baseline in accuracy while using 5.19-6.13 times fewer writes, and up to 9.19 times fewer writes on easier configurations. Budget-matched random and periodic schedules do not recover this gain, isolating the benefit to the action-surprise signal. On a trained closed-loop OpenVLA-OFT 7B panel on LIBERO-Long (n=60 episodes per arm), the gate does not hurt success: AURA-Mem matches the ungated base policy (0.233) and slightly exceeds an always-write KV arm (0.217), while using 7.0 times fewer writes and constant memory. We also instantiate an approximate-information-state value-loss bound as a methodology demonstration; at this scale, the bound is vacuous rather than a guarantee.

Automatic metrics are now central to evaluating text-to-image models, often substituting for human judgment in benchmarking and large-scale filtering. However, it remains unclear whether these metrics truly prioritize semantic correctness or instead favor visually and socially prototypical images learned from biased data distributions. We identify and study prototypicality bias as a systematic failure mode in multimodal evaluation. We introduce a controlled contrastive benchmark ProtoBias (Prototypical Bias), spanning Animals, Objects, and Demography images, where semantically correct but non-prototypical images are paired with subtly incorrect yet prototypical adversarial counterparts. This setup enables a directional evaluation of whether metrics follow textual semantics or default to prototypes. Our results show that widely used metrics, including CLIPScore, PickScore, and VQA-based scores, frequently misrank these pairs, while even LLM-as-Judge systems exhibit uneven robustness in socially grounded cases. Human evaluations consistently favour semantic correctness with larger decision margins. Motivated by these findings, we propose ProtoScore, a robust 7B-parameter metric that substantially reduces failure rates and suppresses misranking, while running at orders of magnitude faster than the inference time of GPT-5, approaching the robustness of much larger closed-source judges.

Financial AI agents often fail for a simple reason: they make users carry the complexity. A user must repeatedly restate goals, risk preferences, portfolio context, past judgments, and shifting market assumptions, while the agent answers, retrieves, acts, and forgets. In finance, this is not just inconvenient. In tasks such as market analysis, copy-trading review, and trade preparation, forgotten context and stale memory can create latency, repeated errors, weak auditability, and unsafe decisions. We propose the interaction-native knowledge harness (InKH), an architecture for financial LLM agents that absorbs complexity into the system. InKH converts user, market, portfolio, and tool events into structured operational knowledge. It uses passive knowledge injection to assemble a bounded working context buffer before the main model step, temporal graph memory for low-latency retrieval, a wiki audit surface for human-readable governance, and background extraction with maturity, decay, and write-time invalidation. We evaluate InKH on a reproducible controlled synthetic benchmark with 24 random seeds, 4 rounds, 80 episodes per round, and 6 baselines, producing 46,080 baseline-conditioned evaluations. InKH achieves mean task quality of 0.815 at 900 ms latency. Compared with agent-driven wiki-walk memory, it reduces latency by 82.95 percent, token cost by 82.29 percent, and stale-knowledge usage by 96.58 percent, while improving quality by 0.108 and traceability by 0.461. Compared with a temporal-graph system without invalidation, it improves quality by 0.050 and reduces stale-memory usage by 96.58 percent with comparable serving cost. The results support a design thesis for financial AI: adoption happens when complexity is absorbed by the system rather than transferred to the user. The benchmark validates architecture-level behavior, not live trading performance.

One-shot Program-of-Thought (PoT) emits a Python program that prints a primitive-action plan; a single invalid action silently invalidates the trajectory. We introduce RePoT (Recoverable PoT): a deterministic verified replay that walks the plan through the environment to its first invalid transition, then one LLM call that resumes from the verified prefix. RePoT costs at most one extra LLM call on the ~14% of problems where PoT fails. RePoT beats PoT by +3 to +11pp across four closed-model configurations on PuzzleZoo-775 and peaks at 96.9% vs 86.3% on gpt-5.4-mini-medium; against the matched-budget PoT-retry baseline, RePoT wins decisively on Gemini (+3.8pp, 95% CI [+2.2,+5.4]), is within sampling noise on GPT-medium and Claude, and loses on GPT-mini -- a capability-scaling pattern we begin to address with Adaptive RePoT, a rule-based dispatcher that routes between suffix repair and a fresh PoT retry based on verified-prefix length (preliminary). We replicate on PlanBench Blocksworld (+1.1 to +11.4pp) and on four open-weights models (+3.3 to +20.0pp on three of four). On Derail-550, our controlled recovery benchmark, every condition with access to checkpoint information clears >=30% on GPT-medium and >=70% on Gemini, vs <=3.1% for error-only feedback -- showing that checkpoint information, not the specific verified-prefix tail, is the load-bearing recovery signal.

Large pre-trained language models have become popular for many applications and form an important backbone of many downstream tasks in natural language processing (NLP). Applying 'explainable artificial intelligence' (XAI) techniques to enrich such models' outputs is considered crucial for assuring their quality and shedding light on their inner workings. However, large language models are trained on a plethora of data containing a variety of biases, such as gender biases, affecting model weights and, potentially, behavior. Currently, it is unclear to what extent such biases also impact model explanations in possibly unfavorable ways. We create a gender-controlled text dataset, GECO, in which otherwise identical sentences appear in male and female forms. This gives rise to ground-truth 'world explanations' for gender classification tasks, enabling the objective evaluation of the correctness of XAI methods. We also provide GECOBench, a rigorous quantitative evaluation framework benchmarking popular XAI methods, applying them to pre-trained language models fine-tuned to different degrees. This allows us to investigate how pre-training induces undesirable bias in model explanations and to what extent fine-tuning can mitigate such explanation bias. We show a clear dependency between explanation performance and the number of fine-tuned layers, where XAI methods are observed to particularly benefit from fine-tuning or complete retraining of embedding layers. Remarkably, this relationship holds for models achieving similar classification performance on the same task. With that, we highlight the utility of the proposed gender-controlled dataset and novel benchmarking approach for research and development of novel XAI methods. All code including dataset generation, model training, evaluation and visualization is available at: https://github.com/braindatalab/gecobench

Large Language Models (LLMs) and Vision-Language Models (VLMs) are increasingly evaluated on table reasoning tasks, but the role of table representation remains under-explored. In practice, the same table content may appear in different structural formats, such as HTML, Markdown, and LaTeX, or as rendered images. However, existing evaluations often let content, format, layout, and modality vary together, making it difficult to isolate representation effects. We introduce TABVERSE, a controlled multimodal table benchmark that aligns the same table content across multiple structural formats and rendered images, with question category and difficulty tags. This design enables systematic evaluation of representation effects while holding table content fixed. We evaluate LLMs and VLMs across three tasks: Question Answering (QA), Structural Understanding Capability (SUC), and Structure Reconstruction (SR). Our results show that representation choice substantially affects table understanding. Models generally perform better with structured text than with rendered images, but the size of this gap depends on the task, model, and format. HTML is often the most robust text format, while row-sensitive structural tasks and syntactically usable LaTeX reconstruction remain challenging. These findings show that table representation is a key factor in reliable table evaluation.

This paper introduces a comprehensive benchmark for evaluating how Large Language Models (LLMs) respond to linguistic shibboleths: subtle linguistic markers that can inadvertently reveal demographic attributes such as gender, social class, or regional background. Through carefully constructed interview simulations using 100 validated question-response pairs, we demonstrate how LLMs systematically penalize certain linguistic patterns, particularly hedging language, despite equivalent content quality. Our benchmark generates controlled linguistic variations that isolate specific phenomena while maintaining semantic equivalence, which enables the precise measurement of demographic bias in automated evaluation systems. We validate our approach along multiple linguistic dimensions, showing that hedged responses receive 25.6% lower ratings on average, and demonstrate the benchmark's effectiveness in identifying model-specific biases. This work establishes a foundational framework for detecting and measuring linguistic discrimination in AI systems, with broad applications to fairness in automated decision-making contexts.

Time Series Language Models (TSLMs) are emerging as unified models for reasoning over continuous signals in natural language. However, long-context retrieval remains a major limitation: existing models are typically trained and evaluated on short sequences, while real-world time-series sensor streams can span millions of datapoints. This mismatch requires precise temporal localization under strict computational constraints, a regime that is not captured by current benchmarks. We introduce TS-Haystack, a long-context temporal retrieval benchmark comprising ten task types across four categories: direct retrieval, temporal reasoning, multi-step reasoning and contextual anomaly. The benchmark uses controlled needle insertion by embedding short activity bouts into longer longitudinal accelerometer recordings, enabling systematic evaluation across context lengths ranging from seconds to 2 hours per sample. We hypothesize that existing TSLM time series encoders overlook temporal granularity as context length increases, creating a task-dependent effect: compression aids classification but impairs retrieval of localized events. Across multiple model and encoding strategies, we observe a consistent divergence between classification and retrieval behavior. Learned latent compression preserves or improves classification accuracy at compression ratios up to 176times, but retrieval performance degrades with context length, incurring in the loss of temporally localized information. These results highlight the importance of architectural designs that decouple sequence length from computational complexity while preserving temporal fidelity.

Offline reinforcement learning (RL) aims at learning a good policy from a batch of collected data, without extra interactions with the environment during training. However, current offline RL benchmarks commonly have a large reality gap, because they involve large datasets collected by highly exploratory policies, and the trained policy is directly evaluated in the environment. In real-world situations, running a highly exploratory policy is prohibited to ensure system safety, the data is commonly very limited, and a trained policy should be well validated before deployment. In this paper, we present a near real-world offline RL benchmark, named NeoRL, which contains datasets from various domains with controlled sizes, and extra test datasets for policy validation. We evaluate existing offline RL algorithms on NeoRL and argue that the performance of a policy should also be compared with the deterministic version of the behavior policy, instead of the dataset reward. The empirical results demonstrate that the tested offline RL algorithms become less competitive to the deterministic policy on many datasets, and the offline policy evaluation hardly helps. The NeoRL suit can be found at http://polixir.ai/research/neorl. We hope this work will shed some light on future research and draw more attention when deploying RL in real-world systems.

Multimodal Deep Learning enhances decision-making by integrating diverse information sources, such as texts, images, audio, and videos. To develop trustworthy multimodal approaches, it is essential to understand how uncertainty impacts these models. We introduce LUMA, a unique benchmark dataset, featuring audio, image, and textual data from 50 classes, for learning from uncertain and multimodal data. It extends the well-known CIFAR 10/100 dataset with audio samples extracted from three audio corpora, and text data generated using the Gemma-7B Large Language Model (LLM). The LUMA dataset enables the controlled injection of varying types and degrees of uncertainty to achieve and tailor specific experiments and benchmarking initiatives. LUMA is also available as a Python package including the functions for generating multiple variants of the dataset with controlling the diversity of the data, the amount of noise for each modality, and adding out-of-distribution samples. A baseline pre-trained model is also provided alongside three uncertainty quantification methods: Monte-Carlo Dropout, Deep Ensemble, and Reliable Conflictive Multi-View Learning. This comprehensive dataset and its tools are intended to promote and support the development and benchmarking of trustworthy and robust multimodal deep learning approaches.

Procedural memory is increasingly used to improve LLM agents on recurring workplace tasks, yet its ability to produce reusable skills remains poorly understood. We introduce AFTER, a benchmark of 382 realistic enterprise tasks spanning six professional roles and 22 procedural skills, designed to evaluate how skills transfer across tasks, roles, and model backbones. The benchmark includes controlled evaluation settings for local improvement, cross-task transfer, cross-role transfer, and cross-model generalization. Experiments show that procedural memory delivers consistent gains in industrial workflows: a single refinement round improves aggregate performance by 3.7-6.7 points, while skills evolved from diverse multi-model execution traces achieve 73.1% cross-model test accuracy, outperforming all single-model trace sources. We further find that some skills generalize broadly across tasks and models, whereas others become specialized to role-specific workflows and lose effectiveness under transfer. These results provide practical guidance for building, evaluating, and deploying procedural memory systems in production agent platforms.

Benchmarks are often used as a standard to understand LLM capabilities in different domains. However, aggregate benchmark scores provide limited insight into compositional skill gaps of LLMs and how to improve them. To make these weaknesses visible, we propose Scaffolded Task Design (STaD) framework. STaD generates controlled variations of benchmark tasks based on the concept of scaffolding, which introduces structured, incremental support in a step-by-step manner. Rather than inspecting failures individually, this approach enables systematic and scalable probing of model behavior by identifying the specific reasoning skill compositions they lack. Treating the LLM as a black box, our experiments on six models of varying sizes reveal multiple failure points in three reasoning benchmarks and highlight each model's unique and distinct skill gaps.

In LLM/VLM agents, prompt privacy risk propagates beyond a single model call because raw user content can flow into retrieval queries, memory writes, tool calls, and logs. Existing de-identification pipelines address document boundaries but not this cross-stage propagation. We propose BodhiPromptShield, a policy-aware framework that detects sensitive spans, routes them via typed placeholders, semantic abstraction, or secure symbolic mapping, and delays restoration to authorized boundaries. Relative to enterprise redaction, this adds explicit propagation-aware mediation and restoration timing as a security variable. Under controlled evaluation on the Controlled Prompt-Privacy Benchmark (CPPB), stage-wise propagation suppresses from 10.7\% to 7.1\% across retrieval, memory, and tool stages; PER reaches 9.3\% with 0.94 AC and 0.92 TSR, outperforming generic de-identification. These are controlled systems results on CPPB rather than formal privacy guarantees or public-benchmark transfer claims. The project repository is available at https://github.com/mabo1215/BodhiPromptShield.git.

Multimodal LLMs can accurately perceive numerical content across modalities yet fail to perform exact multi-digit multiplication when the identical underlying arithmetic problem is presented as numerals, number words, images, or in audio form. Because existing benchmarks often lack systematically paired instances across modalities, it remains difficult to compare genuine arithmetic limits within and across model families. We therefore introduce a controlled multimodal multiplication benchmark that factorially varies digit length, digit sparsity, representation (e.g., numerals vs. number words), and modality (text, rendered images, audio), with paired instances from a reproducible generator. We also define arithmetic load, C, as the product of the total and non-zero digit count as a compact, mechanistically motivated proxy for operation count. Across evaluations, accuracy falls sharply as C grows, often nearing zero by C > 100. Indeed, C remains predictive of performance across modalities and models, with R-squared often > 0.5, nearing the value from more complex measures of arithmetic load that count the number of intermediate arithmetic steps. A separate perception-versus-computation decomposition shows that multimodal degradation is primarily computational rather than perceptual: on matched-perception checks, models are near-perfect (> 99%) across modalities, even when multiplication accuracy drops. Beyond measuring when models fail, we ask which procedures they are predisposed to follow. We introduce a forced-completion loss probe that scores heuristic-specific reasoning prefixes--including columnar multiplication, distributive decomposition, and rounding/compensation. Here, decomposition is favored in both text and vision modalities; heuristic-specific LoRA adapters produce near-orthogonal updates yet degrade accuracy, indicating the base model maintains a well-tuned internal router.

Pretrained language models are commonly aligned with human preferences and downstream tasks via reinforcement finetuning (RFT), which entails maximizing a (possibly learned) reward function using policy gradient algorithms. This work highlights a fundamental optimization obstacle in RFT: we prove that the expected gradient for an input vanishes when its reward standard deviation under the model is small, even if the expected reward is far from optimal. Through experiments on an RFT benchmark and controlled environments, as well as a theoretical analysis, we then demonstrate that vanishing gradients due to small reward standard deviation are prevalent and detrimental, leading to extremely slow reward maximization. Lastly, we explore ways to overcome vanishing gradients in RFT. We find the common practice of an initial supervised finetuning (SFT) phase to be the most promising candidate, which sheds light on its importance in an RFT pipeline. Moreover, we show that a relatively small number of SFT optimization steps on as few as 1% of the input samples can suffice, indicating that the initial SFT phase need not be expensive in terms of compute and data labeling efforts. Overall, our results emphasize that being mindful for inputs whose expected gradient vanishes, as measured by the reward standard deviation, is crucial for successful execution of RFT.

Joint vision-language models have shown great performance over a diverse set of tasks. However, little is known about their limitations, as the high dimensional space learned by these models makes it difficult to identify semantic errors. Recent work has addressed this problem by designing highly controlled probing task benchmarks. Our paper introduces a more scalable solution that relies on already annotated benchmarks. Our method consists of extracting a large set of diverse features from a vision-language benchmark and measuring their correlation with the output of the target model. We confirm previous findings that CLIP behaves like a bag of words model and performs better with nouns and verbs; we also uncover novel insights such as CLIP getting confused by concrete words. Our framework is available at https://github.com/MichiganNLP/Scalable-VLM-Probing and can be used with other multimodal models and benchmarks.

Distinctive Image Captioning (DIC) -- generating distinctive captions that describe the unique details of a target image -- has received considerable attention over the last few years. A recent DIC work proposes to generate distinctive captions by comparing the target image with a set of semantic-similar reference images, i.e., reference-based DIC (Ref-DIC). It aims to make the generated captions can tell apart the target and reference images. Unfortunately, reference images used by existing Ref-DIC works are easy to distinguish: these reference images only resemble the target image at scene-level and have few common objects, such that a Ref-DIC model can trivially generate distinctive captions even without considering the reference images. To ensure Ref-DIC models really perceive the unique objects (or attributes) in target images, we first propose two new Ref-DIC benchmarks. Specifically, we design a two-stage matching mechanism, which strictly controls the similarity between the target and reference images at object-/attribute- level (vs. scene-level). Secondly, to generate distinctive captions, we develop a strong Transformer-based Ref-DIC baseline, dubbed as TransDIC. It not only extracts visual features from the target image, but also encodes the differences between objects in the target and reference images. Finally, for more trustworthy benchmarking, we propose a new evaluation metric named DisCIDEr for Ref-DIC, which evaluates both the accuracy and distinctiveness of the generated captions. Experimental results demonstrate that our TransDIC can generate distinctive captions. Besides, it outperforms several state-of-the-art models on the two new benchmarks over different metrics.

Safe completion requires models to provide useful assistance without enabling harm, but this behavior is difficult to evaluate with isolated prompts. We introduce OpenSafeIntent, a benchmark of controlled prompt-sets that vary intent while holding the underlying task fixed. Each datapoint contains benign, dual-use, and malicious variants of the same task. This design lets us evaluate whether models calibrate assistance across intent shifts, rather than merely appearing safe on average. Across a broad model suite, we find that prompt-level safety hides important failures: models often fail to remain safe across matched intent variants, dual-use behavior is brittle under paraphrase, high-level answers on risky topics are not reliably safe, and responses that reframe ambiguous requests into safer tasks are substantially less likely to cross the safety boundary. Our results suggest that safe completion should be evaluated as intent-calibrated behavior over controlled task variants, not as a single safety-helpfulness tradeoff over independent prompts.

Building performant Vision-Language Models (VLMs) requires carefully curating large-scale training datasets, yet the community lacks systematic benchmarks for evaluating such curation strategies. We introduce DataComp for VLMs (DCVLM), a benchmark for controlled data-centric experiments to improve VLM training. As part of DCVLM, we collect 160 datasets spanning four data types -- image-caption pairs, multimodal interleaved documents, text-only, and instruction-tuning data -- into a corpus of 6T multimodal tokens. DCVLM allows participants to test curation strategies (filtering, mixing, formatting, sampling) across 1B-8B models and 6.25B-200B token budgets. Models are then evaluated on a carefully selected suite of up to 52 downstream benchmarks across 9 domains. We conduct extensive experiments on DCVLM and find that data mixing, not filtering, is key to a high-quality training dataset: instruction-heavy mixtures scale better than caption-heavy ones, with gains widening at larger scales. The resulting dataset, DCVLM-Baseline, enables training an 8B VLM to 63.6% accuracy on our 33-task core suite with 200B training tokens. Compared to FineVision, the state-of-the-art open VLM training dataset, this represents an improvement of +5.4pp. DCVLM and all accompanying artifacts will be made publicly available at https://www.datacomp.ai/dcvlm/.

We study how different Chain-of-Thought (CoT) designs affect the acquisition of the generalizable visual reasoning ability in vision-language models (VLMs). While CoT data, especially long or visual CoT such as "think with image", has been widely used to supervise intermediate reasoning, it remains unclear why specific CoT designs help and which ones truly support generalizable reasoning. To systematically evaluate this, we focus on a controlled maze-solving benchmark where reasoning rules are fully visual, difficulty can be tuned by grid size, and all the intermediate steps can be automatically generated. Using Qwen2.5-VL-7B under a standard SFT-then-RL pipeline, we compare three representative CoT formats: Language CoT, Grounding CoT (with spatial coordinate trajectories), and Visual CoT (with image manipulations). Our experiments reveal that visual and longer CoT mainly accelerate convergence but do not lift the final performance ceiling; concise CoT containing only essential grounding steps outperforms longer traces; and, strikingly, CoT retaining only the minimal grounding results generalizes best across different maze sizes. We further validate these insights on other vision-centric tasks. These findings highlight a "short is long" effect and provide practical guidance for constructing more generalizable SFT datasets for visual reasoning.

Instruction granularity is an important yet poorly controlled variable in language-guided embodied AI. Existing benchmarks typically pair each task with a single static instruction, making it difficult to study how agent behavior changes when the same task is described at different levels of detail. We introduce Mini-BEHAVIOR-Gran, a new benchmark for controlled studies of instruction granularity that extends Mini-BEHAVIOR with multiple instruction variants per task, ranging from high-level goal descriptions to step-by-step guidance. Using this benchmark, we compare four candidate metrics for cross-task granularity quantification: token count, entity count, action-verb count, and planning-width, and find that width correlates most consistently with agent performance. Using width to organize training and evaluation further reveals a non-monotonic U-shaped relationship between instruction granularity and performance, with peaks at both fine and coarse extremes. Further analysis suggests that the coarse-granularity performance rebound is associated with shallow grounding, where agents learn vision-dominant policies.

Multimodal large language models (MLLMs) have achieved remarkable performance on complex tasks with multimodal context. However, it is still understudied whether they exhibit modality preference when processing multimodal contexts. To study this question, we first build a MC\textsuperscript{2} benchmark under controlled evidence conflict scenarios to systematically evaluate modality preference, which is the tendency to favor one modality over another when making decisions based on multimodal conflicting evidence. Our extensive evaluation reveals that all 18 tested MLLMs generally demonstrate clear modality bias, and modality preference can be influenced by external interventions. An in-depth analysis reveals that the preference direction can be captured within the latent representations of MLLMs. Built on this, we propose a probing and steering method based on representation engineering to explicitly control modality preference without additional fine-tuning or carefully crafted prompts. Our method effectively amplifies modality preference toward a desired direction and applies to downstream tasks such as hallucination mitigation and multimodal machine translation, yielding promising improvements.

Alignment in large language models (LLMs) is still largely static: after training, the policy is frozen. DPO, GRPO methods typically imprint one behavior into the weights, leaving little runtime control beyond prompt hacks or expensive re-alignment. We introduce ECLIPTICA, which treats alignment as instruction-driven and runtime-controllable: natural-language alignment instructions act as an explicit behavioral contract (stance, refusal boundary, verbosity) that modulates behavior on the fly under evolving safety requirements, user roles, and governance constraints. We introduce CITA (Contrastive Instruction-Tuned Alignment), combining SFT with contrastive preference optimization under an explicit geometric anchor to a reference model. This yields a stable Riemannian chart and keeps instruction updates within a shared neighborhood, so regimes stay nearby and traversable for reliable switching. To isolate policy switching from ordinary instruction following, we release the ECLIPTICA benchmark: 3000 controlled cases (300 prompts x 10 instruction types) where the user request is fixed and only the alignment instruction changes. On Llama-3.1-8B across five suites (ECLIPTICA, TruthfulQA, Conditional Safety, Length Control, LITMUS), CITA reaches 86.7% instruction-alignment efficiency, beating DPO (56.1%), GRPO (36.1%), and PPO (20.4%).

Probabilistic super-resolution of high-dimensional spatial fields using diffusion models is often computationally prohibitive due to the cost of operating directly in pixel space. We propose PODiff, a structured conditional generative framework that performs diffusion in a fixed, variance-ordered Proper Orthogonal Decomposition (POD) coefficient space, exploiting the orthogonality of POD modes to impose an interpretable, variance-ordered latent geometry. This design enables efficient ensemble generation, preserves dominant spatial structure, and yields spatially interpretable, well-calibrated uncertainty at substantially lower computational cost. We evaluate PODiff on sea surface temperature downscaling over the West Australian coast and on a controlled advection-diffusion benchmark. PODiff achieves reconstruction accuracy comparable to pixel-space diffusion while requiring significantly less memory and producing more reliable uncertainty estimates than deterministic and Monte Carlo Dropout baselines.

In semi-supervised semantic segmentation, existing studies have shown promising results in academic settings with controlled splits of benchmark datasets. However, the potential benefits of leveraging significantly larger sets of unlabeled images remain unexplored. In real-world scenarios, abundant unlabeled images are often available from online sources (web-scraped images) or large-scale datasets. However, these images may have different distributions from those of the target dataset, a situation known as out-of-distribution (OOD). Using these images as unlabeled data in semi-supervised learning can lead to inaccurate pseudo-labels, potentially misguiding network training. In this paper, we propose a new semi-supervised semantic segmentation framework with an open-vocabulary segmentation model (SemiOVS) to effectively utilize unlabeled OOD images. Extensive experiments on Pascal VOC and Context datasets demonstrate two key findings: (1) using additional unlabeled images improves the performance of semi-supervised learners in scenarios with few labels, and (2) using the open-vocabulary segmentation (OVS) model to pseudo-label OOD images leads to substantial performance gains. In particular, SemiOVS outperforms existing PrevMatch and SemiVL methods by +3.5 and +3.0 mIoU, respectively, on Pascal VOC with a 92-label setting, achieving state-of-the-art performance. These findings demonstrate that our approach effectively utilizes abundant unlabeled OOD images for semantic segmentation tasks. We hope this work can inspire future research and real-world applications. The code is available at https://github.com/wooseok-shin/SemiOVS

The real world is non-stationary and infinitely complex, requiring intelligent agents to learn continually without the prohibitive cost of retraining from scratch. While online continual learning offers a framework for this setting, learning new information often interferes with previously acquired knowledge, causes forgetting and degraded generalization. To address this, we propose Abstraction-Augmented Training (AAT), a loss-level modification encouraging models to capture the latent relational structure shared across examples. By jointly optimizing over concrete instances and their abstract representations, AAT introduces a memory-efficient inductive bias that stabilizes learning in strictly online data streams, eliminating the need for a replay buffer. To capture the multi-faceted nature of abstraction, we introduce and evaluate AAT on two benchmarks: a controlled relational dataset where abstraction is realized through entity masking, and a narrative dataset where abstraction is expressed through shared proverbs. Our results show that AAT achieves performance comparable to or exceeding strong experience replay (ER) baselines, despite requiring zero additional memory and only minimal changes to the training objective. This work highlights structural abstraction as a powerful, memory-free alternative to ER.

Recent large-scale language models (LLMs) with long Chain-of-Thought reasoning-such as DeepSeek-R1-have achieved impressive results on Olympiad-level mathematics benchmarks. However, they often rely on a narrow set of strategies and struggle with problems that require a novel way of thinking. To systematically investigate these limitations, we introduce OMEGA-Out-of-distribution Math Problems Evaluation with 3 Generalization Axes-a controlled yet diverse benchmark designed to evaluate three axes of out-of-distribution generalization, inspired by Boden's typology of creativity: (1) Exploratory-applying known problem solving skills to more complex instances within the same problem domain; (2) Compositional-combining distinct reasoning skills, previously learned in isolation, to solve novel problems that require integrating these skills in new and coherent ways; and (3) Transformative-adopting novel, often unconventional strategies by moving beyond familiar approaches to solve problems more effectively. OMEGA consists of programmatically generated training-test pairs derived from templated problem generators across geometry, number theory, algebra, combinatorics, logic, and puzzles, with solutions verified using symbolic, numerical, or graphical methods. We evaluate frontier (or top-tier) LLMs and observe sharp performance degradation as problem complexity increases. Moreover, we fine-tune the Qwen-series models across all generalization settings and observe notable improvements in exploratory generalization, while compositional generalization remains limited and transformative reasoning shows little to no improvement. By isolating and quantifying these fine-grained failures, OMEGA lays the groundwork for advancing LLMs toward genuine mathematical creativity beyond mechanical proficiency.

Recent advances in pretraining 3D point cloud encoders (e.g., Point-BERT, Point-MAE) have produced powerful models, whose abilities are typically evaluated on geometric or semantic tasks. At the same time, topological descriptors have been shown to provide informative summaries of a shape's multiscale structure. In this paper we pose the question whether topological information can be derived from features produced by 3D encoders. To address this question, we first introduce DONUT, a synthetic benchmark with controlled topological complexity, and propose FILTR (Filtration Transformer), a learnable framework to predict persistence diagrams directly from frozen encoders. FILTR adapts a transformer decoder to treat diagram generation as a set prediction task. Our analysis on DONUT reveals that existing encoders retain only limited global topological signals, yet FILTR successfully leverages information produced by these encoders to approximate persistence diagrams. Our approach enables, for the first time, data-driven extraction of persistence diagrams from raw point clouds through an efficient learnable feed-forward mechanism.

Parameter-efficient fine-tuning (PEFT) has become a standard approach for adapting large pre-trained models. Amongst PEFT methods, low-rank adaptation (LoRA) has achieved notable success. However, recent studies have highlighted its limitations compared against full-rank alternatives, particularly when applied to multimodal and large language models. In this work, we present a quantitative comparison amongst full-rank and low-rank PEFT methods using a synthetic matrix approximation benchmark with controlled spectral properties. Our results confirm that LoRA struggles to approximate matrices with relatively flat spectrums or high frequency components -- signs of high effective ranks. To this end, we introduce KRAdapter, a novel PEFT algorithm that leverages the Khatri-Rao product to produce weight updates, which, by construction, tends to produce matrix product with a high effective rank. We demonstrate performance gains with KRAdapter on vision-language models up to 1B parameters and on large language models up to 8B parameters, particularly on unseen common-sense reasoning tasks. In addition, KRAdapter maintains the memory and compute efficiency of LoRA, making it a practical and robust alternative to fine-tune billion-scale parameter models.

The development of clinical natural language processing (NLP) systems is severely hampered by the sensitive nature of medical records, which restricts data sharing under stringent privacy regulations, particularly in France and the broader European Union. To address this gap, we introduce PARHAF, a large open-source corpus of clinical documents in French. PARHAF comprises expert-authored clinical reports describing realistic yet entirely fictitious patient cases, making it anonymous and freely shareable by design. The corpus was developed using a structured protocol that combined clinician expertise with epidemiological guidance from the French National Health Data System (SNDS), ensuring broad clinical coverage. A total of 104 medical residents across 18 specialties authored and peer-reviewed the reports following predefined clinical scenarios and document templates. The corpus contains 7394 clinical reports covering 5009 patient cases across a wide range of medical and surgical specialties. It includes a general-purpose component designed to approximate real-world hospitalization distributions, and four specialized subsets that support information-extraction use cases in oncology, infectious diseases, and diagnostic coding. Documents are released under a CC-BY open license, with a portion temporarily embargoed to enable future benchmarking under controlled conditions. PARHAF provides a valuable resource for training and evaluating French clinical language models in a fully privacy-preserving setting, and establishes a replicable methodology for building shareable synthetic clinical corpora in other languages and health systems.

Multilingual benchmarks rarely test reasoning over culturally grounded premises: translated datasets keep English-centric scenarios, while culture-first datasets often lack control over the reasoning required. We propose Macaron, a template-first benchmark that factorizes reasoning type and cultural aspect across question languages. Using 100 language-agnostic templates that cover 7 reasoning types, 22 cultural aspects, native annotators create scenario-aligned English and local-language multiple-choice questions and systematically derived True/False questions. Macaron contains 11,862 instances spanning 20 countries/cultural contexts, 10 scripts, and 20 languages (including low-resource ones like Amharic, Yoruba, Zulu, Kyrgyz, and some Arabic dialects). In zero-shot evaluation of 21 multilingual LLMs, reasoning-mode models achieve the strongest performance and near-parity between English and local languages, while open-weight models degrade substantially in local languages and often approach chance on T/F tasks. Culture-grounded mathematical and counting templates are consistently the hardest. The data can be accessed here https://huggingface.co/datasets/AlaaAhmed2444/Macaron.

Evaluating the scientific discovery capabilities of large language model based agents, particularly how they cope with varying environmental complexity and utilize prior knowledge, requires specialized benchmarks currently lacking in the landscape. To address this gap, we introduce PhysGym, a novel benchmark suite and simulation platform for rigorously assessing LLM-based scientific reasoning in interactive physics environments. PhysGym's primary contribution lies in its sophisticated control over the level of prior knowledge provided to the agent. This allows researchers to dissect agent performance along axes including the complexity of the problem and the prior knowledge levels. The benchmark comprises a suite of interactive simulations, where agents must actively probe environments, gather data sequentially under constraints and formulate hypotheses about underlying physical laws. PhysGym provides standardized evaluation protocols and metrics for assessing hypothesis accuracy and model fidelity. We demonstrate the benchmark's utility by presenting results from baseline LLMs, showcasing its ability to differentiate capabilities based on varying priors and task complexity.

Recommender-system research for Vietnamese remains limited by the absence of a public, well-documented hotel interaction resource. Building such a resource is challenging for three reasons: cross-platform hotel names must be reconciled before interactions are comparable; quality must be audited with reproducible metrics rather than ad hoc cleaning; and public release must preserve privacy while remaining benchmarkable under realistic cold-start conditions. We introduce ViHoRec, a quality-controlled Vietnamese hotel recommendation dataset of 18{,}267 interactions between 6{,}832 users and 560 hotels, crawled from Booking.com, Traveloka, and Ivivu. Our contributions are: (i) a reproducible construction pipeline with cross-platform entity resolution and quantitative quality control; (ii) a privacy-preserving release with HMAC pseudonyms; and (iii) a public cold-start benchmark with temporal leave-last-one-out split, data-centric ablations, and dependency-free baselines. On the public split, learned models degrade sharply for users with short histories (BPR-MF Recall@10: 0.065 vs. 0.120), while UserKNN remains strongest overall, establishing ViHoRec as a sparse, cold-start-dominated testbed for low-resource recommendation. All data are publicly available at https://github.com/MinhNguyenDS/ViHoRec.

Stereo image and video generation, stereo geometry estimation, and condition-controlled view synthesis require paired data in which the variables that determine binocular geometry -- camera baseline, intrinsics, scene depth, and camera motion -- are known and controllable. Existing stereo resources provide subsets of these variables, but resources commonly used for stereo generation evaluation do not, to our knowledge, provide scene-paired, calibrated multi-baseline right-view ground truth with jointly recorded intrinsics, dense metric depth, and per-frame poses in a single controlled source. We introduce StereoGenBench, a synthetic Unreal Engine benchmark designed to make baseline-regime sensitivity and target-camera consistency measurable under matched scene content. Each scene is rendered with a rigid six-camera lateral array, yielding up to 15 calibrated view pairs; adjacent baselines are sampled from inter-pupillary to wide-baseline regimes; focal length is sampled independently; and every view is released with RGB, metric depth, intrinsics, per-pair baselines, and per-frame poses. The splits include two evaluation families for narrow and wide baseline regimes and a train-only family for broader all-pairs coverage. We release the dataset, evaluation code, reference results, Croissant metadata, and generation code/configuration for extension with compatible assets. The dataset is available at https://huggingface.co/datasets/stereo-dataset/stereo-dataset

Multimodal Large Language Models (MLLMs) excel at recognizing individual visual elements and reasoning over simple linear diagrams. However, when faced with complex topological structures involving branching paths, converging flows, and cyclic dependencies, their reasoning capabilities degrade sharply, even on tasks as basic as counting endpoints. Existing benchmarks fail to probe this gap, focusing on semantic comprehension rather than structural reasoning. We introduce ReactBench, a benchmark that reveals fundamental limitations in structural reasoning through chemical reaction diagrams. These real-world scientific diagrams offer an ideal testbed because they naturally span diverse structures from linear chains to cyclic graphs, while requiring both precise local recognition and coherent global reasoning. Our benchmark comprises 1,618 expert-annotated QA pairs across four hierarchical task dimensions. Extensive evaluation across 17 MLLMs reveals a significant performance gap exceeding 30% between anchor-based tasks and holistic structural reasoning tasks. Controlled ablations confirm this bottleneck lies in reasoning, not perception. These findings expose a fundamental deficit in structural understanding and establish directions for advancing visual reasoning.

Multimodal models are increasingly deployed to solve tasks collaboratively with humans or other artificial agents. Existing benchmarks show that these models possess many of the required component capabilities, but the conditions that coincide in collaboration, including time pressure, information asymmetry, and imperfect communication, are usually studied in isolation. We introduce GPTNT, a benchmark built on the cooperative video game Keep Talking and Nobody Explodes, in which two agents must coordinate to defuse procedurally generated bomb puzzles against a live countdown. One agent can see and manipulate the bomb but does not have the defusal instructions; the other has the instructions but cannot see or manipulate the bomb. Neither agent can succeed alone: success requires effective and efficient communication. Unlike turn-based proxies, GPTNT requires agents to act asynchronously and communicate in real time. GPTNT is designed to separate collaboration from reliance on memorized solutions: the instruction manual, the partner, or both can be withheld to isolate what a model derives in the moment from what it already knows. We show that GPTNT poses a substantial challenge for state-of-the-art systems: none of the closed- or open-source models we test defuses a single bomb in real time, a bar that human players clear. Through controlled experiments, we identify critical weaknesses in state tracking, efficient action under time pressure, ambiguity handling, and error recovery. We release GPTNT as a benchmark for collaborative performance that current evaluations leave unmeasured. Because it runs on the real game, GPTNT benefits from procedural generation and inherits a living modding community, allowing the benchmark to evolve as models improve rather than being solved once and retired.

Benchmarks increasingly guide deployment, procurement and scientific screening, yet a score supports only the response it records, not necessarily the deployment action. We introduce deployment-complete benchmarking, which tests whether benchmark evidence determines a deployment action. A benchmark is complete for a claim exactly when the action is constant on each evidence fiber; mixed fibers expose missing deployment information, and completion curves quantify the evidence required to resolve ambiguity. In controlled response spaces, benchmark-channel conformal coverage of 94.98% transferred poorly to an unmeasured deployment channel (10.07%), whereas response-rank intervals achieved 94.91% coverage; even zero benchmark error certified only 45.4% of candidates at the largest residual size. Public audits revealed incompleteness, including 97.9% mixed Tox21 fibers and zero median certifiable fraction in main Matbench and JARVIS audits. In held-out replays, certify-then-acquire reduced false decisions from 1.19% to 0.027% in Tox21 and from 20.3% to 0.128% in JARVIS, while changing model choice and identifying deployment-relevant probes. Deployment-ready benchmarks should report evidence, supported actions, ambiguity and completion cost rather than scores alone.

AI for scientific discovery is entering an agentic era, where protein-engineering systems are expected to prioritize future wet-lab experiments rather than merely fit static measurements. We introduce TadA-Bench, a million-variant wet-lab replay benchmark from 31 TadA directed-evolution rounds for future-round discovery toward agentic protein engineering. TadA-Bench preserves the campaign chronology and defines a fixed-data replay task: given earlier experimental rounds, models rank variants that appear only in later rounds. It provides aligned DNA, RNA, and protein views, and uses Seq2Graph, a graph-based label-unification pipeline, to reconcile noisy enrichment measurements into consistent cross-round activity labels. Random-split controls show strong interpolation, but future-round ranking and finite-budget candidate selection are much weaker. Controlled analyses suggest that evolutionary coverage is more informative than local data density, positioning TadA-Bench as a reproducible wet-lab replay substrate for future-round discovery toward agentic protein engineering; the data and code are released on Hugging Face and GitHub.

Reinforcement learning (RL) trained language model agents with tool access are increasingly deployed in coding assistants, research tools, and autonomous systems. We introduce the Reward Hacking Benchmark (RHB), a suite of multi-step tasks requiring sequential tool operations with naturalistic shortcut opportunities such as skipping verification steps, inferring answers from task-adjacent metadata, or tampering with evaluation-relevant functions. RHB supports independent and chained task regimes, where chain length acts as a proxy for longer-horizon agent behavior. We evaluate 13 frontier models from OpenAI, Anthropic, Google, and DeepSeek. Exploit rates range from 0% (Claude Sonnet 4.5) to 13.9% (DeepSeek-R1-Zero), varying sharply by post-training style. A controlled sibling comparison (DeepSeek-V3 vs. DeepSeek-R1-Zero) shows RL post-training is associated with substantially higher reward hacking (0.6% vs. 13.9%), with consistent gaps across all four task families. We identify six exploit categories and find that 72% of reward hacking episodes include explicit chain-of-thought rationale, suggesting models often frame exploits as legitimate problem-solving. Simple environmental hardening reduces exploit rates by 5.7 percentage points (87.7% relative) without degrading task success. Models with near-zero exploit rates on standard tasks show elevated rates on harder variants, suggesting that production-aligned post-training appears to suppress reward hacking only below a complexity threshold where honest solutions remain tractable.

Video prediction is increasingly viewed as a path toward generalizable world models, yet it remains unclear whether these systems learn underlying causal structure or merely exploit superficial visual correlations for future prediction. We introduce CRONOS, an intervention-based benchmark designed to evaluate counterfactual physical consistency: whether a model's predictions of physical events respond appropriately to controlled changes in the visual input, such as variations of scene context, viewpoint, object appearance, and object category. Built in a photorealistic Unreal Engine environment, CRONOS enables controlled, high-fidelity generation of videos across diverse scenes and dynamics. In contrast to previous benchmarks, CRONOS systematically intervenes on four key factors - viewpoint, scene, object category, and object appearance - while keeping the underlying physical event type, such as a collision, occlusion, or fall, fixed. Our evaluation of recent open-source video generators reveals substantial failures in counterfactual physical consistency: prediction quality for the same physical event type is affected by appearance, environment, and, particularly by viewpoint changes. CRONOS provides a controlled and reproducible testbed for diagnosing how the quality of generated videos changes for different interventions, establishing a concrete target for developing models that perform consistently across changes of multiple conditions. The dataset and code are available at our project page.

Code large language models (Code LLMs) have made significant progress in code generation by translating natural language descriptions into functional code; however, real-world applications often demand stricter adherence to detailed requirements such as coding style, line count, and structural constraints, beyond mere correctness. To address this, the paper introduces forward and backward constraints generation to improve the instruction-following capabilities of Code LLMs in controlled code generation, ensuring outputs align more closely with human-defined guidelines. The authors further present IFEvalCode, a multilingual benchmark comprising 1.6K test samples across seven programming languages (Python, Java, JavaScript, TypeScript, Shell, C++, and C#), with each sample featuring both Chinese and English queries. Unlike existing benchmarks, IFEvalCode decouples evaluation into two metrics: correctness (Corr.) and instruction-following (Instr.), enabling a more nuanced assessment. Experiments on over 40 LLMs reveal that closed-source models outperform open-source ones in controllable code generation and highlight a significant gap between the models' ability to generate correct code versus code that precisely follows instructions.

Persistent AI assistants, such as OpenClaw, accumulate large collections of related memories over long-term interactions. As these memories grow, they may reinforce one another, diverge across contexts, or directly conflict, making correct assistance depend on memory relations rather than isolated recall. Existing long-term memory benchmarks rarely probe how agents preserve and utilize such relations during downstream tasks. To address this gap, we introduce SubtleMemory, a benchmark for fine-grained relational memory discrimination in long-running AI agents. SubtleMemory constructs relation-controlled latent semantic artifacts whose variants instantiate complementary, nuanced, or contradictory relations, and embeds them into realistic user-agent histories, requiring agents to recover distributed relational structures during later queries and instructions. The benchmark contains 1,522 evaluation instances over 10 long histories, grounded in 1,090 relation-controlled memory-variant sets and spanning user-related and non-user-related queries. Evaluating six standalone memory systems, two Claw-style agents with native memory modules, and three Claw-style agents with plugin memory modules, we find that current systems remain weak on fine-grained relational memory discrimination. We further introduce diagnostic protocols that reveal distinct capability profiles across memory preservation, retrieval, and downstream reasoning stages.

We introduce BigO(Bench), a novel coding benchmark designed to evaluate the capabilities of generative language models in understanding and generating code with specified time and space complexities. This benchmark addresses the gap in current evaluations that often overlook the ability of models to comprehend and produce code constrained by computational complexity. BigO(Bench) includes tooling to infer the algorithmic complexity of any Python function from profiling measurements, including human- or LLM-generated solutions. BigO(Bench) also includes of set of 3,105 coding problems and 1,190,250 solutions from Code Contests annotated with inferred (synthetic) time and space complexity labels from the complexity framework, as well as corresponding runtime and memory footprint values for a large set of input sizes. We present results from evaluating multiple state-of-the-art language models on this benchmark, highlighting their strengths and weaknesses in handling complexity requirements. In particular, token-space reasoning models are unrivaled in code generation but not in complexity understanding, hinting that they may not generalize well to tasks for which no reward was given at training time.

Knowledge graphs automatically constructed from text are increasingly used in real-world applications. However, their inherent noise, fragmentation, and semantic inconsistencies significantly affect the performance of Graph Neural Networks (GNNs) on downstream tasks. Assessing their performance and robustness remains difficult, as it is often unclear whether observed results stem from the learning model or from the quality of the constructed graph itself. In this work, we introduce a dual-purpose benchmark designed to jointly evaluate (i) the performance of GNNs on noisy, text-derived graphs and (ii) the effectiveness of graph construction methods on a downstream task. The benchmark is built in the biomedical domain from a single textual corpus and includes two automatically constructed graphs generated using different extraction methods, alongside a high-quality reference graph curated by experts that serves as an upper performance bound. This design enables controlled comparison of construction methods and systematic evaluation of GNN robustness through semi-supervised node classification. We further provide a standardized, reproducible, and extensible evaluation framework, facilitating the integration of new graph extraction methods and learning models.

Solving topological grid puzzles requires reasoning over global spatial invariants such as connectivity, loop closure, and region symmetry and remains challenging for even the most powerful large language models (LLMs). To study these abilities under controlled settings, we introduce TopoBench, a benchmark of six puzzle families across three difficulty levels. We evaluate strong reasoning LLMs on TopoBench and find that even frontier models solve fewer than one quarter of hard instances, with two families nearly unsolved. To investigate whether these failures stem from reasoning limitations or from difficulty extracting and maintaining spatial constraints, we annotate 750 chain of thought traces with an error taxonomy that surfaces four candidate causal failure modes, then test them with targeted interventions simulating each error type. These interventions show that certain error patterns like premature commitment and constraint forgetting have a direct impact on the ability to solve the puzzle, while repeated reasoning is a benign effect of search. Finally we study mitigation strategies including prompt guidance, cell-aligned grid representations and tool-based constraint checking, finding that the bottleneck lies in extracting constraints from spatial representations and not in reasoning over them. Code and data are available at github.com/mayug/topobench-benchmark.

Vision-Language Models (VLMs) have advanced across multimodal benchmarks but still show clear gaps in ordinal number understanding, i.e., the ability to track relative positions and generalize to large indices. We present OrdinalBench, a diagnostic benchmark that standardizes ordinal number understanding as an evaluation task for VLMs. The core task is N-th object identification, defined by a starting reference and traversal rule. Task difficulty is controlled along three axes: (i) ordinal magnitude, from small numbers to extreme cases up to 300; (ii) arrangement complexity, from single loops to maze-like paths; and (iii) object count. The benchmark provides 39,000 question-answer pairs, each annotated with a ground-truth reasoning trajectory and balanced across difficulty levels for controlled large-scale testing. Beyond answer-only evaluation, our framework requires models to generate structured stepwise traces of the counting process and provides an open evaluation toolkit that measures both final accuracy and step-level path consistency. Zero-shot evaluations of GPT-5, Gemini 2.5 Flash Lite, Qwen2.5-VL, InternVL3.5, and Molmo reveal sharp degradation under large-ordinal and complex-path conditions, highlighting weak generalization despite strong scores on standard multimodal tasks. By framing ordinal number understanding as a core target, OrdinalBench provides a reproducible benchmark and diagnostic framework for developing VLMs with stronger sequential reasoning. All data and code are available at https://ordinalbench.github.io/

Reasoning-focused Question Answering (QA) has advanced rapidly with Large Language Models (LLMs), yet high-quality benchmarks for low-resource languages remain scarce. Persian, spoken by roughly 130 million people, lacks a comprehensive open-domain resource for evaluating reasoning-capable QA systems. We introduce PARSE, the first open-domain Persian reasoning QA benchmark, containing 10,800 questions across Boolean, multiple-choice, and factoid formats, with diverse reasoning types, difficulty levels, and answer structures. The benchmark is built via a controlled LLM-based generation pipeline and validated through human evaluation. We also ensure linguistic and factual quality through multi-stage filtering, annotation, and consistency checks. We benchmark multilingual and Persian LLMs under multiple prompting strategies and show that Persian prompts and structured prompting (CoT for Boolean/multiple-choice; few-shot for factoid) improve performance. Fine-tuning further boosts results, especially for Persian-specialized models. These findings highlight how PARSE supports both fair comparison and practical model adaptation. PARSE fills a critical gap in Persian QA research and provides a strong foundation for developing and evaluating reasoning-capable LLMs in low-resource settings.

We introduce CompareBench, a benchmark for evaluating visual comparison reasoning in vision-language models (VLMs), a fundamental yet understudied skill. CompareBench consists of 1000 QA pairs across four tasks: quantity (600), temporal (100), geometric (200), and spatial (100). It is derived from two auxiliary datasets that we constructed: TallyBench (2000 counting images with QA) and HistCaps (515 historical images with bilingual captions). We evaluate both closed-source APIs (OpenAI, Gemini, Claude) and open-source models (Qwen2.5-VL and Qwen3-VL series). Results show clear scaling trends but also reveal critical limitations: even the strongest models consistently fail at temporal ordering and spatial relations, and they often make mistakes in basic counting and geometric comparisons that are trivial for humans. These findings demonstrate that visual comparison remains a systematic blind spot for current VLMs. By providing controlled, diverse, and diagnostic evaluation, CompareBench establishes a foundation for advancing more reliable multimodal reasoning.

We introduce MCUBench, a benchmark featuring over 100 YOLO-based object detection models evaluated on the VOC dataset across seven different MCUs. This benchmark provides detailed data on average precision, latency, RAM, and Flash usage for various input resolutions and YOLO-based one-stage detectors. By conducting a controlled comparison with a fixed training pipeline, we collect comprehensive performance metrics. Our Pareto-optimal analysis shows that integrating modern detection heads and training techniques allows various YOLO architectures, including legacy models like YOLOv3, to achieve a highly efficient tradeoff between mean Average Precision (mAP) and latency. MCUBench serves as a valuable tool for benchmarking the MCU performance of contemporary object detectors and aids in model selection based on specific constraints.

Many social and environmental phenomena are associated with macroscopic changes in the built environment, captured by satellite imagery on a global scale and with daily temporal resolution. While widely used for prediction, these images and especially image sequences remain underutilized for causal inference, especially in the context of randomized controlled trials (RCTs), where causal identification is established by design. In this paper, we develop and compare a set of general tools for analyzing Conditional Average Treatment Effects (CATEs) from temporal satellite data that can be applied to any RCT where geographical identifiers are available. Through a simulation study, we analyze different modeling strategies for estimating CATE in sequences of satellite images. We find that image sequence representation models with more parameters generally yield a greater ability to detect heterogeneity. To explore the role of model and data choice in practice, we apply the approaches to two influential RCTs -- Banerjee et al. (2015), a poverty study in Cusco, Peru, and Bolsen et al. (2014), a water conservation experiment in Georgia, USA. We benchmark our image sequence models against image-only, tabular-only, and combined image-tabular data sources, summarizing practical implications for investigators in a multivariate analysis. Land cover classifications over satellite images facilitate interpretation of what image features drive heterogeneity. We also show robustness to data and model choice of satellite-based generalization of the RCT results to larger geographical areas outside the original. Overall, this paper shows how satellite sequence data can be incorporated into the analysis of RCTs, and provides evidence about the implications of data, model, and evaluation metric choice for causal analysis.

Natural Language Inference (NLI) tasks involving temporal inference remain challenging for pre-trained language models (LMs). Although various datasets have been created for this task, they primarily focus on English and do not address the need for resources in other languages. It is unclear whether current LMs realize the generalization capacity for temporal inference across languages. In this paper, we present Jamp, a Japanese NLI benchmark focused on temporal inference. Our dataset includes a range of temporal inference patterns, which enables us to conduct fine-grained analysis. To begin the data annotation process, we create diverse inference templates based on the formal semantics test suites. We then automatically generate diverse NLI examples by using the Japanese case frame dictionary and well-designed templates while controlling the distribution of inference patterns and gold labels. We evaluate the generalization capacities of monolingual/multilingual LMs by splitting our dataset based on tense fragments (i.e., temporal inference patterns). Our findings demonstrate that LMs struggle with specific linguistic phenomena, such as habituality, indicating that there is potential for the development of more effective NLI models across languages.

We introduce REAL, a benchmark and framework for multi-turn agent evaluations on deterministic simulations of real-world websites. REAL comprises high-fidelity, deterministic replicas of 11 widely-used websites across domains such as e-commerce, travel, communication, and professional networking. We also release a benchmark consisting of 112 practical tasks that mirror everyday complex user interactions requiring both accurate information retrieval and state-changing actions. All interactions occur within this fully controlled setting, eliminating safety risks and enabling robust, reproducible evaluation of agent capability and reliability. Our novel evaluation framework combines programmatic checks of website state for action-based tasks with rubric-guided LLM-based judgments for information retrieval. The framework supports both open-source and proprietary agent systems through a flexible evaluation harness that accommodates black-box commands within browser environments, allowing research labs to test agentic systems without modification. Our empirical results show that frontier language models achieve at most a 41% success rate on REAL, highlighting critical gaps in autonomous web navigation and task completion capabilities. Our framework supports easy integration of new tasks, reproducible evaluation, and scalable post-training data generation, marking a significant step forward in evaluating and advancing agent capabilities.

Large Language Model (LLM) agents increasingly rely on external memory systems to accumulate experience across tasks. Yet nearly all existing approaches, from graph-structured memories to reflective insight stores, access memory through fixed, hand-designed heuristics. We argue that this static view of memory is a core bottleneck for agentic learning because optimal memory behavior is fundamentally context-dependent. The early stages of the tasks, benefit from minimal retrieval because memory is sparse; recurring goal types benefit from plan reuse rather than generic nearest-neighbor lookup; stuck agents benefit from re-retrieval with alternative queries; and across long task streams, the memory store itself must be consolidated and pruned to remain useful. We present Memory as a Controlled Process (MemCon), a framework that models memory operations as a Markov Decision Process and learns an online policy that adaptively decides when, what, and how much to retrieve, when to inject a distilled plan, and when to consolidate or forget. MemCon is backend-agnostic: it wraps any existing memory implementation, learns from task-by-task binary feedback with no pretraining and no additional LLM calls, and uses a lightweight tabular contextual bandit with UCB exploration that converges within tens of tasks. Across 6 benchmarks, 3 agent frameworks, and 3 LLM backbones, MemCon consistently outperforms multiple memory baselines by up to 15.2 points in task success while reducing token consumption by 5--20%.

We introduce iTRIALSPACE, a programmable evaluation framework for controlled assessment of lung CT models. Standard benchmarks are static retrospective collections that entangle lesion size, lobe prevalence, anatomy, and acquisition context, making it difficult to determine what structurally drives model accuracy. iTRIALSPACE addresses this limitation by composing real clinical CTs and lesion profiles into controlled virtual lesion trials through a four-stage pipeline: multidataset nodule profiling, explicit trial specification, anatomy-aware mask insertion, and ControlNet-conditioned CT synthesis. The framework is built on a unified 54-attribute nodule-profile dataset spanning 13,140 annotated nodules from seven public CT sources and instantiated as 13 trial modes. We evaluate iTRIALSPACE in a 55,469-sample Virtual Lesion Study spanning three medical VLMs, four spatialguidance conditions, and three clinical tasks. Across all 13 modes, the synthetic substrate remains within the real-to-real FID baseline, and synthetic performance rankings transfer strongly to real clinical data (ρ = 0.93, p < 10^{-15}). Controlled trial modes expose findings unavailable to fixed-distribution benchmarks, including shortcut-driven size prediction collapse under lobe-equalized sampling and hostto-donor variance ratios of 8.9x and 3.3x in twin-cross analysis. These results position iTRIALSPACE as an auditable evaluation infrastructure for controlled, falsifiable testing beyond static retrospective benchmarks.

Despite strong performance on existing benchmarks, it remains unclear whether large language models can reason over genuinely novel scientific information. Most evaluations score end-to-end RAG pipelines, where reasoning is confounded with retrieval and toolchain choices, and the signal is further contaminated by parametric memorization and open-web volatility. We introduce DeR2, a controlled deep-research sandbox that isolates document-grounded reasoning while preserving core difficulties of deep search: multi-step synthesis, denoising, and evidence-based conclusion making. DeR2 decouples evidence access from reasoning via four regimes--Instruction-only, Concepts (gold concepts without documents), Related-only (only relevant documents), and Full-set (relevant documents plus topically related distractors)--yielding interpretable regime gaps that operationalize retrieval loss vs. reasoning loss and enable fine-grained error attribution. To prevent parametric leakage, we apply a two-phase validation that requires parametric failure without evidence while ensuring oracle-concept solvability. To ensure reproducibility, each instance provides a frozen document library (drawn from 2023-2025 theoretical papers) with expert-annotated concepts and validated rationales. Experiments across a diverse set of state-of-the-art foundation models reveal substantial variation and significant headroom: some models exhibit mode-switch fragility, performing worse with the Full-set than with Instruction-only, while others show structural concept misuse, correctly naming concepts but failing to execute them as procedures.

Re-rendering an existing video from a novel camera viewpoint requires the output to follow the prescribed camera trajectory while preserving the appearance and dynamics of the original scene across every frame. Existing methods rely on per-frame pose embeddings, noisy point-cloud renderings, or implicit learned correspondences, none of which provides an explicit, temporally continuous link between source and target pixels. We propose Track2View, which conditions a video diffusion transformer on paired 3D point tracks: sparse trajectories of scene points projected into both the source and target camera views. These tracks provide explicit spatiotemporal correspondences that are temporally continuous by construction, encoding what content should appear where and when. At the core of Track2View is a dual-view track conditioner that transfers visual context from source to target view through parameter-free geometric operations and learned temporal aggregation, ensuring generalization to arbitrary camera trajectories without memorizing specific motions. We further introduce a data curation pipeline that extracts one-to-one track correspondences by running a 3D point tracker on temporally concatenated multi-camera view pairs. On a 400-video benchmark spanning static and dynamic scenes, Track2View achieves state-of-the-art results across visual quality, view synchronization, and camera accuracy, reducing rotation error by 30-65% and translation error by 61-72% relative to leading baselines. Project page is available at this https URL: https://qjizhi.github.io/track2view

Evaluating whether vision-language models (VLMs) reason consistently across representations is challenging because modality comparisons are typically confounded by task differences and asymmetric information. We introduce SEAM, a benchmark that pairs semantically equivalent inputs across four domains that have existing standardized textual and visual notations. By employing distinct notation systems across modalities, in contrast to OCR-based image-text pairing, SEAM provides a rigorous comparative assessment of the textual-symbolic and visual-spatial reasoning capabilities of VLMs. Across 21 contemporary models, we observe systematic modality imbalance: vision frequently lags language in overall performance, despite the problems containing semantically equivalent information, and cross-modal agreement is relatively low. Our error analysis reveals two main drivers: textual perception failures from tokenization in domain notation and visual perception failures that induce hallucinations. We also show that our results are largely robust to visual transformations. SEAM establishes a controlled, semantically equivalent setting for measuring and improving modality-agnostic reasoning.

Multimodal Large Language Models (MLLMs) have been increasingly used as automatic evaluators-a paradigm known as MLLM-as-a-Judge. However, their reliability and vulnerabilities to biases remain underexplored. We find that many MLLM judges fail to reliably integrate key visual or textual cues, yielding unreliable evaluations when evidence is missing or mismatched, and exhibiting instability under semantically irrelevant perturbations. To address this, we systematically define Compositional Bias in MLLM-as-a-Judge systems and introduce MM-JudgeBias, a benchmark for evaluating it. MM-JudgeBias introduces controlled perturbations across Query, Image, and Response, and evaluates model behavior via two complementary metrics: Bias-Deviation (BD) for sensitivity and Bias-Conformity (BC) for stability. Our dataset of over 1,800 curated and refined multimodal samples, drawn from 29 source benchmarks, enables a fine-grained diagnosis of nine bias types across diverse tasks and domains. Experiments on 26 state-of-the-art MLLMs reveal systematic modality neglect and asymmetric evaluation tendencies, underscoring the need for more reliable judges.

Fine-tuning multilingual ASR models like Whisper for low-resource languages often improves read speech but degrades spontaneous audio performance, a phenomenon we term studio-bias. To diagnose this mismatch, we introduce Vividh-ASR, a complexity-stratified benchmark for Hindi and Malayalam across four tiers: studio, broadcast, spontaneous, and synthetic noise. Through a controlled study of learning-rate timing and curriculum ordering, we find that early large parameter updates improve global WER by 12 absolute points, while a hard-to-easy curriculum adds gains for spontaneous speech. These findings motivate reverse multi-stage fine-tuning (R-MFT), a training recipe that enables a parameter-efficient 244M Whisper model to match or exceed conventionally fine-tuned 769M counterparts. Representational analysis via CKA and SVD reveals effective schedules concentrate adaptation in the decoder, preserving the pre-trained encoder's acoustic geometry. We release the benchmark and models.

Scientific reasoning in materials science requires integrating multimodal experimental evidence with underlying physical theory. Existing benchmarks make it difficult to assess whether incorporating visual experimental data during post-training improves mechanism-grounded explanation reasoning beyond text-only supervision. We introduce MATRIX, a multimodal benchmark for materials science reasoning that evaluates foundational theory, research-level reasoning, and the interpretation of real experimental artifacts across multiple characterization modalities. Using MATRIX as a controlled diagnostic, we isolate the effect of visual grounding by comparing post-training on structured materials science text alone with post-training that incorporates paired experimental images. Despite using relatively small amounts of multimodal data, visual supervision improves experimental interpretation by 10-25% and yields 5-16% gains on text-only scientific reasoning tasks. Our results demonstrate that these improvements rely on correct image-text alignment during post-training, highlighting cross-modal representational transfer. We also observe consistent improvements on ScienceQA and PubMedQA, demonstrating that the benefits of structured multimodal post-training extend beyond materials science. The MATRIX dataset is available at https://huggingface.co/datasets/radical-ai/MATRIX and the model at https://huggingface.co/radical-ai/MATRIX-PT.

Mathematical reasoning remains challenging for LLMs due to complex logic and the need for precise computation. Existing methods enhance LLM reasoning by synthesizing datasets through problem rephrasing, but face issues with generation quality and problem complexity. To address this, we propose to extract structural information with generated problem-solving code from mathematical reasoning and guide data generation with structured solutions. Applied to MATH and GSM8K, our approach produces 39K problems with labeled intermediate steps and a 6.1K-problem benchmark of higher difficulty. Results on our benchmark show that model performance declines as reasoning length increases. Additionally, we conducted fine-tuning experiments using the proposed training data on a range of LLMs, and the results validate the effectiveness of our dataset. We hope the proposed method and dataset will contribute to future research in enhancing LLM reasoning capabilities. Our code and data are available at https://github.com/OpenCausaLab/StructuralGeneration.

Feed-forward 3D reconstruction has advanced rapidly, but current models remain unreliable in UAV photogrammetric acquisition. We argue that this failure is caused not only by appearance-domain shift, but also by UAV-specific camera-geometry variations, especially oblique views and HFOV-height ambiguity. Existing UAV datasets mainly emphasize scene diversity and provide limited coverage of camera configurations, which restricts robustness evaluation and UAV-domain adaptation. To address this gap, we introduce UAVFF3D, a geometry-aware real-synthetic benchmark for feed-forward UAV 3D reconstruction. UAVFF3D contains more than 170k real UAV images and more than 370k synthetic images rendered from high-quality textured 3D models, covering diverse HFOVs, flight altitudes, viewing directions, and acquisition patterns. It also includes a controlled HFOV-height test subset for diagnosing projection-geometry ambiguity. We further propose an evaluation protocol that jointly assesses camera-geometry estimation and dense scene reconstruction under a shared global alignment, avoiding the bias caused by separate camera and geometry alignments. Experiments on representative feed-forward reconstruction models show that UAVFF3D-based domain adaptation consistently improves camera and geometry estimation, reducing Ray Error by up to 84.2%, Pose ATE by up to 76.0%, and Chamfer Distance by up to 41.1%. In oblique scenes, adaptation reduces the oblique-nadir rotation gap by up to 90.7%. Under HFOV-height ambiguity, it improves robustness across HFOV-height configurations and yields more stable performance across HFOV settings. Incorporating camera priors further improves reconstruction under UAV-specific acquisition geometries. The dataset and evaluation code are available at https://github.com/yanxian-ll/UAVFF3D .

Long video question answering (Long-Video QA) increasingly relies on agentic tool use to retrieve evidence from long videos. In realistic settings, this process often requires multi-hop retrieval, where agents must iteratively gather multiple discontinuous evidence clips. However, existing long-video benchmarks are largely static: they rarely enforce strict multi-hop retrieval and typically lack a standardized evidence-access interface, making it difficult to separate failures in retrieval planning from those in answer generation. To address this gap, we introduce LongVidSearch, a benchmark for evaluating agentic multi-hop evidence retrieval planning in long videos under standardized access constraints. LongVidSearch enforces retrieval necessity: a Hop-k question requires exactly k necessary evidence clips, and removing any single clip renders the question unsolvable. The benchmark contains 3,000 questions over 447 long videos (average length 26 minutes), covering four reasoning categories: State Mutation, Causal Inference, Global Summary, and Visual Tracking, with 2-hop, 3-hop, and 4-hop evidence requirements. To ensure fair and controlled evaluation, all agents interact with LongVidSearch through a unified tool interface, which fixes the retrieval backend and isolates the agent's ability to formulate queries and plan iterative retrieval. In addition to answer accuracy, we measure tool-call cost to analyze the accuracy-efficiency trade-off under identical access conditions. We evaluate VideoAgent-style QA agents with multiple backbone LLMs using three-judge majority voting. GPT-5 achieves the highest accuracy (42.43), outperforming Gemini 3 Pro (30.97) and GPT-4o (19.20), yet remaining below 50 %, highlighting the difficulty of multi-hop retrieval planning. With gold evidence clips, performance becomes near-perfect, confirming retrieval planning as the primary bottleneck.

We present DOCFORGE-BENCH, the first unified zero-shot benchmark for document forgery detection, evaluating 14 methods across eight datasets spanning text tampering, receipt forgery, and identity document manipulation. Unlike fine-tuning-oriented evaluations such as ForensicHub [Du et al., 2025], DOCFORGE-BENCH applies all methods with their published pretrained weights and no domain adaptation -- a deliberate design choice that reflects the realistic deployment scenario where practitioners lack labeled document training data. Our central finding is a pervasive calibration failure invisible under single-threshold protocols: methods achieve moderate Pixel-AUC (>=0.76) yet near-zero Pixel-F1. This AUC-F1 gap is not a discrimination failure but a score-distribution shift: tampered regions occupy only 0.27-4.17% of pixels in document images -- an order of magnitude less than in natural image benchmarks -- making the standard tau=0.5 threshold catastrophically miscalibrated. Oracle-F1 is 2-10x higher than fixed-threshold Pixel-F1, confirming that calibration, not representation, is the bottleneck. A controlled calibration experiment validates this: adapting a single threshold on N=10 domain images recovers 39-55% of the Oracle-F1 gap, demonstrating that threshold adaptation -- not retraining -- is the key missing step for practical deployment. Overall, no evaluated method works reliably out-of-the-box on diverse document types, underscoring that document forgery detection remains an unsolved problem. We further note that all eight datasets predate the era of generative AI editing; benchmarks covering diffusion- and LLM-based document forgeries represent a critical open gap on the modern attack surface.

Multi-hop reasoning over financial disclosures is often a retrieval problem before it becomes a reasoning or generation problem: relevant facts are dispersed across sections, filings, companies, and years, and LLMs often expend excessive tokens navigating noisy context. Without precise Knowledge Graph (KG)-guided selection of relevant context, even strong reasoning models either fail to answer or consume excessive tokens, whereas KG-linked evidence enables models to focus their reasoning on composing already retrieved facts. We present FinReflectKG - MultiHop, a benchmark built on FinReflectKG, a temporally indexed financial KG that links audited triples to source chunks from S&P 100 filings (2022-2024). Mining frequent 2-3 hop subgraph patterns across sectors (via GICS taxonomy), we generate financial analyst style questions with exact supporting evidence from the KG. A two-phase pipeline first creates QA pairs via pattern-specific prompts, followed by a multi-criteria quality control evaluation to ensure QA validity. We then evaluate three controlled retrieval scenarios: (S1) precise KG-linked paths; (S2) text-only page windows centered on relevant text spans; and (S3) relevant page windows with randomizations and distractors. Across both reasoning and non-reasoning models, KG-guided precise retrieval yields substantial gains on the FinReflectKG - MultiHop QA benchmark dataset, boosting correctness scores by approximately 24 percent while reducing token utilization by approximately 84.5 percent compared to the page window setting, which reflects the traditional vector retrieval paradigm. Spanning intra-document, inter-year, and cross-company scopes, our work underscores the pivotal role of knowledge graphs in efficiently connecting evidence for multi-hop financial QA. We also release a curated subset of the benchmark (555 QA Pairs) to catalyze further research.

Medical report interpretation plays a crucial role in healthcare, enabling both patient-facing explanations and effective information flow across clinical systems. While recent vision-language models (VLMs) and large language models (LLMs) have demonstrated general document understanding capabilities, there remains a lack of standardized benchmarks to assess structured interpretation quality in medical reports. We introduce MedRepBench, a comprehensive benchmark built from 1,900 de-identified real-world Chinese medical reports spanning diverse departments, patient demographics, and acquisition formats. The benchmark is designed primarily to evaluate end-to-end VLMs for structured medical report understanding. To enable controlled comparisons, we also include a text-only evaluation setting using high-quality OCR outputs combined with LLMs, allowing us to estimate the upper-bound performance when character recognition errors are minimized. Our evaluation framework supports two complementary protocols: (1) an objective evaluation measuring field-level recall of structured clinical items, and (2) an automated subjective evaluation using a powerful LLM as a scoring agent to assess factuality, interpretability, and reasoning quality. Based on the objective metric, we further design a reward function and apply Group Relative Policy Optimization (GRPO) to improve a mid-scale VLM, achieving up to 6% recall gain. We also observe that the OCR+LLM pipeline, despite strong performance, suffers from layout-blindness and latency issues, motivating further progress toward robust, fully vision-based report understanding.

Dynamic graph learning methods have recently emerged as powerful tools for modelling relational data evolving through time. However, despite extensive benchmarking efforts, it remains unclear whether current Temporal Graph Neural Networks (TGNNs) effectively capture core temporal patterns such as periodicity, cause-and-effect, and long-range dependencies. In this work, we introduce the Temporal Graph Reasoning Benchmark (T-GRAB), a comprehensive set of synthetic tasks designed to systematically probe the capabilities of TGNNs to reason across time. T-GRAB provides controlled, interpretable tasks that isolate key temporal skills: counting/memorizing periodic repetitions, inferring delayed causal effects, and capturing long-range dependencies over both spatial and temporal dimensions. We evaluate 11 temporal graph learning methods on these tasks, revealing fundamental shortcomings in their ability to generalize temporal patterns. Our findings offer actionable insights into the limitations of current models, highlight challenges hidden by traditional real-world benchmarks, and motivate the development of architectures with stronger temporal reasoning abilities. The code for T-GRAB can be found at: https://github.com/alirezadizaji/T-GRAB.

Dataset contamination, where evaluation datasets overlap with pre-training corpora, inflates performance metrics and undermines the reliability of model evaluations. Measuring dataset contamination thus becomes essential to ensure that performance evaluations genuinely reflect a model's ability to generalize to unseen data, rather than relying on memorized examples. To address this problem, we propose Kernel Divergence Score (KDS), a novel method that evaluates dataset contamination by computing the divergence between the kernel similarity matrix of sample embeddings, before and after fine-tuning on the benchmark dataset. Leveraging the insight that fine-tuning affects unseen samples more significantly than seen ones, KDS provides a reliable measure of contamination. Through extensive experiments on controlled contamination scenarios, KDS demonstrates a near-perfect correlation with contamination levels and outperforms existing baselines. Additionally, we perform comprehensive ablation studies to analyze the impact of key design choices, providing deeper insights into the components and effectiveness of KDS. These ablations highlight the importance of leveraging fine-grained kernel-based information and confirm the reliability of the proposed framework across diverse datasets and settings. Code is released in https://github.com/deeplearning-wisc/kernel-divergence-score.

Developing autonomous home robots controlled by natural language has long been a pursuit of human. While advancements in large language models (LLMs) and embodied intelligence make this goal closer, several challenges persist: the lack of a unified benchmark for more complex robot tasks, limited evaluation methods and metrics, data incompatibility between LLMs and mobile manipulation trajectories. To address these issues, we introduce Embodied Mobile Manipulation in Open Environments (EMMOE), which requires agents to interpret user instructions and execute long-horizon everyday tasks in continuous space. EMMOE seamlessly integrates high-level and low-level embodied tasks into a unified framework, along with three new metrics for more diverse assessment. Additionally, we collect EMMOE-100, which features in various task attributes, detailed process annotations, re-plans after failures, and two sub-datasets for LLM training. Furthermore, we design HomieBot, a sophisticated agent system consists of LLM with Direct Preference Optimization (DPO), light weighted navigation and manipulation models, and multiple error detection mechanisms. Finally, we demonstrate HomieBot's performance and the evaluation of different models and policies.

Omni-modal large language models (OLMs) redefine human-machine interaction by natively integrating audio, vision, and text. However, existing OLM benchmarks remain anchored to static, accuracy-centric tasks, leaving a critical gap in assessing social interactivity, the fundamental capacity to navigate dynamic cues in natural dialogues. To this end, we propose SocialOmni, a comprehensive benchmark that operationalizes the evaluation of this conversational interactivity across three core dimensions: (i) speaker separation and identification (who is speaking), (ii) interruption timing control (when to interject), and (iii) natural interruption generation (how to phrase the interruption). SocialOmni features 2,000 perception samples and a quality-controlled diagnostic set of 209 interaction-generation instances with strict temporal and contextual constraints, complemented by controlled audio-visual inconsistency scenarios to test model robustness. We benchmarked 12 leading OLMs, which uncovers significant variance in their social-interaction capabilities across models. Furthermore, our analysis reveals a pronounced decoupling between a model's perceptual accuracy and its ability to generate contextually appropriate interruptions, indicating that understanding-centric metrics alone are insufficient to characterize conversational social competence. More encouragingly, these diagnostics from SocialOmni yield actionable signals for bridging the perception-interaction divide in future OLMs.

Nucleotide sequences constitute the fundamental genetic basis of biological systems, rendering viral genomic analysis critical for biomedical advancement. Despite progress in biological foundation models, specifically nucleotide foundation models (NFMs), the field lacks a unified standard for viral genomics to facilitate community development and enforce biosecurity constraints. To address this, we introduce ViroBench, the first comprehensive and large-scale benchmark specifically designed for NFMs in viral settings. ViroBench evaluates models across two critical dimensions: biological understanding and latent biosecurity risk, covering 18 diverse scenarios within 4 task types. Extensive evaluation of 66 NFMs across diverse architectures yields three critical conclusions. Firstly, NFMs exhibit a performance degradation in biological understanding under phylogenetic and temporal shifts, indicating weak extrapolation capabilities. Secondly, generation tasks reveal a decoupling between statistical likelihood and biological functional validity, posing latent biosecurity risks. Thirdly, controlled ablation studies reveal that taxonomic diversity in pretraining data outweighs parameter scale. Specifically, a lightweight baseline trained on diverse data achieves a 67.5% performance gain over its original model. Overall, ViroBench provides interpretable, diagnostic evaluations and a reproducible measurement framework for future research on viral nucleotide foundation models. The datasets and code are publicly available at https://github.com/QIANJINYDX/ViroBench.

While neural language models often perform surprisingly well on natural language understanding (NLU) tasks, their strengths and limitations remain poorly understood. Controlled synthetic tasks are thus an increasingly important resource for diagnosing model behavior. In this work we focus on story understanding, a core competency for NLU systems. However, the main synthetic resource for story understanding, the bAbI benchmark, lacks such a systematic mechanism for controllable task generation. We develop Dyna-bAbI, a dynamic framework providing fine-grained control over task generation in bAbI. We demonstrate our ideas by constructing three new tasks requiring compositional generalization, an important evaluation setting absent from the original benchmark. We tested both special-purpose models developed for bAbI as well as state-of-the-art pre-trained methods, and found that while both approaches solve the original tasks (>99% accuracy), neither approach succeeded in the compositional generalization setting, indicating the limitations of the original training data. We explored ways to augment the original data, and found that though diversifying training data was far more useful than simply increasing dataset size, it was still insufficient for driving robust compositional generalization (with <70% accuracy for complex compositions). Our results underscore the importance of highly controllable task generators for creating robust NLU systems through a virtuous cycle of model and data development.

Patent prior-art retrieval becomes especially challenging when relevant disclosures cross technological boundaries. Existing benchmarks lack explicit domain partitions, making it difficult to assess how retrieval systems cope with such shifts. We introduce DAPFAM, a family-level benchmark with explicit IN-domain and OUT-domain partitions defined by a new IPC3 overlap scheme. The dataset contains 1,247 query families and 45,336 target families aggregated at the family level to reduce international redundancy, with citation based relevance judgments. We conduct 249 controlled experiments spanning lexical (BM25) and dense (transformer) backends, document and passage level retrieval, multiple query and document representations, aggregation strategies, and hybrid fusion via Reciprocal Rank Fusion (RRF). Results reveal a pronounced domain gap: OUT-domain performance remains roughly five times lower than IN-domain across all configurations. Passage-level retrieval consistently outperforms document-level, and dense methods provide modest gains over BM25, but none close the OUT-domain gap. Document-level RRF yields strong effectiveness efficiency trade-offs with minimal overhead. By exposing the persistent challenge of cross-domain retrieval, DAPFAM provides a reproducible, compute-aware testbed for developing more robust patent IR systems. The dataset is publicly available on huggingface at https://huggingface.co/datasets/datalyes/DAPFAM_patent.

Automated fish documentation processes are in the near future expected to play an essential role in sustainable fisheries management and for addressing challenges of overfishing. In this paper, we present a novel and publicly available dataset named AutoFish designed for fine-grained fish analysis. The dataset comprises 1,500 images of 454 specimens of visually similar fish placed in various constellations on a white conveyor belt and annotated with instance segmentation masks, IDs, and length measurements. The data was collected in a controlled environment using an RGB camera. The annotation procedure involved manual point annotations, initial segmentation masks proposed by the Segment Anything Model (SAM), and subsequent manual correction of the masks. We establish baseline instance segmentation results using two variations of the Mask2Former architecture, with the best performing model reaching an mAP of 89.15%. Additionally, we present two baseline length estimation methods, the best performing being a custom MobileNetV2-based regression model reaching an MAE of 0.62cm in images with no occlusion and 1.38cm in images with occlusion. Link to project page: https://vap.aau.dk/autofish/.

Deep learning in computer vision has achieved great success with the price of large-scale labeled training data. However, exhaustive data annotation is impracticable for each task of all domains of interest, due to high labor costs and unguaranteed labeling accuracy. Besides, the uncontrollable data collection process produces non-IID training and test data, where undesired duplication may exist. All these nuisances may hinder the verification of typical theories and exposure to new findings. To circumvent them, an alternative is to generate synthetic data via 3D rendering with domain randomization. We in this work push forward along this line by doing profound and extensive research on bare supervised learning and downstream domain adaptation. Specifically, under the well-controlled, IID data setting enabled by 3D rendering, we systematically verify the typical, important learning insights, e.g., shortcut learning, and discover the new laws of various data regimes and network architectures in generalization. We further investigate the effect of image formation factors on generalization, e.g., object scale, material texture, illumination, camera viewpoint, and background in a 3D scene. Moreover, we use the simulation-to-reality adaptation as a downstream task for comparing the transferability between synthetic and real data when used for pre-training, which demonstrates that synthetic data pre-training is also promising to improve real test results. Lastly, to promote future research, we develop a new large-scale synthetic-to-real benchmark for image classification, termed S2RDA, which provides more significant challenges for transfer from simulation to reality. The code and datasets are available at https://github.com/huitangtang/On_the_Utility_of_Synthetic_Data.

The rapid advancement of generative AI has revolutionized image creation, enabling high-quality synthesis from text prompts while raising critical challenges for media authenticity. We present Ai-GenBench, a novel benchmark designed to address the urgent need for robust detection of AI-generated images in real-world scenarios. Unlike existing solutions that evaluate models on static datasets, Ai-GenBench introduces a temporal evaluation framework where detection methods are incrementally trained on synthetic images, historically ordered by their generative models, to test their ability to generalize to new generative models, such as the transition from GANs to diffusion models. Our benchmark focuses on high-quality, diverse visual content and overcomes key limitations of current approaches, including arbitrary dataset splits, unfair comparisons, and excessive computational demands. Ai-GenBench provides a comprehensive dataset, a standardized evaluation protocol, and accessible tools for both researchers and non-experts (e.g., journalists, fact-checkers), ensuring reproducibility while maintaining practical training requirements. By establishing clear evaluation rules and controlled augmentation strategies, Ai-GenBench enables meaningful comparison of detection methods and scalable solutions. Code and data are publicly available to ensure reproducibility and to support the development of robust forensic detectors to keep pace with the rise of new synthetic generators.

Imagery collected from outdoor visual environments is often degraded due to the presence of dense smoke or haze. A key challenge for research in scene understanding in these degraded visual environments (DVE) is the lack of representative benchmark datasets. These datasets are required to evaluate state-of-the-art vision algorithms (e.g., detection and tracking) in degraded settings. In this paper, we address some of these limitations by introducing the first realistic hazy image benchmark, from both aerial and ground view, with paired haze-free images, and in-situ haze density measurements. This dataset was produced in a controlled environment with professional smoke generating machines that covered the entire scene, and consists of images captured from the perspective of both an unmanned aerial vehicle (UAV) and an unmanned ground vehicle (UGV). We also evaluate a set of representative state-of-the-art dehazing approaches as well as object detectors on the dataset. The full dataset presented in this paper, including the ground truth object classification bounding boxes and haze density measurements, is provided for the community to evaluate their algorithms at: https://a2i2-archangel.vision. A subset of this dataset has been used for the ``Object Detection in Haze'' Track of CVPR UG2 2022 challenge at http://cvpr2022.ug2challenge.org/track1.html.