Daily Papers

Vision-and-Language Navigation (VLN) aims to navigate to the target location by following a given instruction. Unlike existing methods focused on predicting a more accurate action at each step in navigation, in this paper, we make the first attempt to tackle a long-ignored problem in VLN: narrowing the gap between Success Rate (SR) and Oracle Success Rate (OSR). We observe a consistently large gap (up to 9%) on four state-of-the-art VLN methods across two benchmark datasets: R2R and REVERIE. The high OSR indicates the robot agent passes the target location, while the low SR suggests the agent actually fails to stop at the target location at last. Instead of predicting actions directly, we propose to mine the target location from a trajectory given by off-the-shelf VLN models. Specially, we design a multi-module transformer-based model for learning compact discriminative trajectory viewpoint representation, which is used to predict the confidence of being a target location as described in the instruction. The proposed method is evaluated on three widely-adopted datasets: R2R, REVERIE and NDH, and shows promising results, demonstrating the potential for more future research.

Vision-language models (VLMs) have been widely-applied in ground-based vision-language navigation (VLN). However, the vast complexity of outdoor aerial environments compounds data acquisition challenges and imposes long-horizon trajectory planning requirements on Unmanned Aerial Vehicles (UAVs), introducing novel complexities for aerial VLN. To address these challenges, we propose a data-efficient Open-world aerial Vision-Language Navigation (i.e., OpenVLN) framework, which could execute language-guided flight with limited data constraints and enhance long-horizon trajectory planning capabilities in complex aerial environments. Specifically, we reconfigure a reinforcement learning framework to optimize the VLM for UAV navigation tasks, which can efficiently fine-tune VLM by using rule-based policies under limited training data. Concurrently, we introduce a long-horizon planner for trajectory synthesis that dynamically generates precise UAV actions via value-based rewards. To the end, we conduct sufficient navigation experiments on the TravelUAV benchmark with dataset scaling across diverse reward settings. Our method demonstrates consistent performance gains of up to 4.34% in Success Rate, 6.19% in Oracle Success Rate, and 4.07% in Success weighted by Path Length over baseline methods, validating its deployment efficacy for long-horizon UAV navigation in complex aerial environments.

Agentic red-teaming benchmarks report whether an injected agent was compromised as a single bit: the attack succeeded, or it did not. We argue that this binary attack-success rate discards the information a defender most needs, namely how harmful the resulting action was. We introduce an action-graded harm rubric that scores an agent's tool-call trajectory on a seven-level ordinal scale (L0 to L6) according to whether the executed action was reversible, whether it crossed scope to reach another party, and whether it expanded privilege. We compute the scale two ways: a deterministic oracle that reads the trajectory and the attacker's stated goal, and a panel of three frontier language-model judges that read a tag-free account of the same trajectory. Across four victim models and two defenses on the AgentDojo workspace suite, severity grading exposes three cases the binary metric hides, including a defense that reports a zero attack-success rate while still permitting an externally visible cross-scope leak through an unfiltered tool. The judge panel reproduces the oracle with high ordinal agreement (Krippendorff's alpha = 0.91) but shares systematic blind spots that we characterize, most notably a failure to recognize escalation chains. Unlike prior work that provides harm taxonomies, harmful-task completion tests, execution-level safety benchmarks, or severity-aware simulation, our contribution is a reusable, trace-grounded severity instrument applied to the actual actions recorded in existing red-team logs. All code, prompts, and per-episode logs are released.

Recent work has proposed stateful defense models (SDMs) as a compelling strategy to defend against a black-box attacker who only has query access to the model, as is common for online machine learning platforms. Such stateful defenses aim to defend against black-box attacks by tracking the query history and detecting and rejecting queries that are "similar" and thus preventing black-box attacks from finding useful gradients and making progress towards finding adversarial attacks within a reasonable query budget. Recent SDMs (e.g., Blacklight and PIHA) have shown remarkable success in defending against state-of-the-art black-box attacks. In this paper, we show that SDMs are highly vulnerable to a new class of adaptive black-box attacks. We propose a novel adaptive black-box attack strategy called Oracle-guided Adaptive Rejection Sampling (OARS) that involves two stages: (1) use initial query patterns to infer key properties about an SDM's defense; and, (2) leverage those extracted properties to design subsequent query patterns to evade the SDM's defense while making progress towards finding adversarial inputs. OARS is broadly applicable as an enhancement to existing black-box attacks - we show how to apply the strategy to enhance six common black-box attacks to be more effective against current class of SDMs. For example, OARS-enhanced versions of black-box attacks improved attack success rate against recent stateful defenses from almost 0% to to almost 100% for multiple datasets within reasonable query budgets.

This paper shows that self-supervised visual pre-training from real-world images is effective for learning motor control tasks from pixels. We first train the visual representations by masked modeling of natural images. We then freeze the visual encoder and train neural network controllers on top with reinforcement learning. We do not perform any task-specific fine-tuning of the encoder; the same visual representations are used for all motor control tasks. To the best of our knowledge, this is the first self-supervised model to exploit real-world images at scale for motor control. To accelerate progress in learning from pixels, we contribute a benchmark suite of hand-designed tasks varying in movements, scenes, and robots. Without relying on labels, state-estimation, or expert demonstrations, we consistently outperform supervised encoders by up to 80% absolute success rate, sometimes even matching the oracle state performance. We also find that in-the-wild images, e.g., from YouTube or Egocentric videos, lead to better visual representations for various manipulation tasks than ImageNet images.

Activation oracles aim to make the activations of other models legible to humans and yield promising results compared to white-box interpretability techniques. However, uncertainty quantification (UQ) for the natural-language outputs of such activation oracles is so far understudied. Here, we investigate 6 different methods for estimating the confidence of activation oracles and evaluate how well-calibrated their confidence scores are. Our experiments on 6,000 samples per oracle (varying verbalizer and context prompts) reveal that bootstrap mode frequency is the best-calibrated method among those tested (ECE 5.7% vs. 25.5% for the answer-word log-probability on Qwen3-8B; 10.3% vs. 13.1% on Qwen3.6-27B), and that the log-prob baseline can serve as a fast triage signal at a fraction of the cost. Code and the patched trainer are available at https://github.com/federicotorrielli/probabilistic_activation_oracles.

Benchmarks are widely used to evaluate task completion by Large Language Models (LLMs), but this approach has accumulated construction-validity problems, and a passing score may not show whether the requested task was delivered. We study both problems. In a controlled code-as-spec setup, two production Copilot CLI agents (claude-opus-4.7, gpt-5.5) re-implement a React Fluent-UI data table in Angular as a reusable library under a hidden 222-test Playwright oracle across 18 runs and three oracle-availability conditions. Alongside the score, we run a mechanical library audit and check each verdict with a no-op ablation. Without the oracle, the library is present but unfinished, revealed by scores. With the oracle in the loop, the score reaches near-perfect, but from a demo holding the tested behavior directly, the library left dead or absent. We call this building to the test; the broader disposition behind both we call validation self-awareness. The agent does not, on its own, validate what it ships as a user would. Prevalence remains an open question across other agents, signals, and model families. Beyond benchmark scores, dispositions like validation self-awareness merit research attention.

Benchmarks for LLM-generated GPU kernels (KernelBench, TritonBench, GEAK) score correctness through fixed-shape, small-sample allclose-style checks. The number of inputs varies between benchmarks. The shape, dtype, and tolerance are fixed for each kernel. We test that oracle empirically. We construct a controlled corpus of 24 Triton and CPU stand-in kernels (15 correct controls and 9 LLM-style buggy variants seeded with documented transcription errors) and re-evaluate it under op-schema-aware seeded fuzzing with a high-precision (fp64) CPU reference and per-(op, dtype) absolute tolerances. The seeded oracle flags 9 of 9 buggy kernels and passes 15 of 15 correct controls, at zero precision cost on controls. We extend the corpus to 26 ops (adding a flash-attention pair) and re-run the same protocol on five GPU classes (RTX 3060, A10, L40S, A100 SXM4, H100 NVL). The verdicts are identical across all five GPUs: 10 of 10 illusions caught and 16 of 16 controls clean. The corpus result is about LLM-style transcription bugs that the allclose-on-one-shape oracle certifies as correct, not about the bug rate of any specific deployed LLM. Every flagged failure replays byte-for-byte from a stored seed.

There is increasing evidence that question-answering (QA) systems with Large Language Models (LLMs), which employ a knowledge graph/semantic representation of an enterprise SQL database (i.e. Text-to-SPARQL), achieve higher accuracy compared to systems that answer questions directly on SQL databases (i.e. Text-to-SQL). Our previous benchmark research showed that by using a knowledge graph, the accuracy improved from 16% to 54%. The question remains: how can we further improve the accuracy and reduce the error rate? Building on the observations of our previous research where the inaccurate LLM-generated SPARQL queries followed incorrect paths, we present an approach that consists of 1) Ontology-based Query Check (OBQC): detects errors by leveraging the ontology of the knowledge graph to check if the LLM-generated SPARQL query matches the semantic of ontology and 2) LLM Repair: use the error explanations with an LLM to repair the SPARQL query. Using the chat with the data benchmark, our primary finding is that our approach increases the overall accuracy to 72% including an additional 8% of "I don't know" unknown results. Thus, the overall error rate is 20%. These results provide further evidence that investing knowledge graphs, namely the ontology, provides higher accuracy for LLM powered question answering systems.

We study whether demonstration-curation metrics that detect defective training episodes also improve the downstream behavior-cloning policy that trains on the curated data. On a contact-rich LIBERO pick-and-place benchmark with a controlled structural defect (early gripper release during the carry phase), we find that the two quantities are sharply decoupled. The metric with the highest defect-detection AUROC (0.804) produces the worst curated policy (13.3% task success), while a metric with a substantially lower AUROC (0.638) produces a policy that nearly matches the oracle trained on ground-truth clean data (90.0% vs. 93.3%). We further show that five of the seven metrics we evaluate exploit episode length as a trivial proxy for the defect label, a confound that inflates reported AUROCs to near-perfect values and disappears once episode length is controlled. Across all conditions, the contaminated baseline succeeds on only 3.3% of rollouts, and the two best curation methods close this to within 3 percentage points of the 93.3% oracle ceiling. Our results argue that curation methods should be evaluated by the policy they produce, not the defects they flag, and that any curation benchmark must control for episode length before reporting detection accuracy. We release the testbed, all metric implementations, and the evaluation pipeline.

The adoption of Large Language Models (LLMs) for code generation in data science offers substantial potential for enhancing tasks such as data manipulation, statistical analysis, and visualization. However, the effectiveness of these models in the data science domain remains underexplored. This paper presents a controlled experiment that empirically assesses the performance of four leading LLM-based AI assistants-Microsoft Copilot (GPT-4 Turbo), ChatGPT (o1-preview), Claude (3.5 Sonnet), and Perplexity Labs (Llama-3.1-70b-instruct)-on a diverse set of data science coding challenges sourced from the Stratacratch platform. Using the Goal-Question-Metric (GQM) approach, we evaluated each model's effectiveness across task types (Analytical, Algorithm, Visualization) and varying difficulty levels. Our findings reveal that all models exceeded a 50% baseline success rate, confirming their capability beyond random chance. Notably, only ChatGPT and Claude achieved success rates significantly above a 60% baseline, though none of the models reached a 70% threshold, indicating limitations in higher standards. ChatGPT demonstrated consistent performance across varying difficulty levels, while Claude's success rate fluctuated with task complexity. Hypothesis testing indicates that task type does not significantly impact success rate overall. For analytical tasks, efficiency analysis shows no significant differences in execution times, though ChatGPT tended to be slower and less predictable despite high success rates. This study provides a structured, empirical evaluation of LLMs in data science, delivering insights that support informed model selection tailored to specific task demands. Our findings establish a framework for future AI assessments, emphasizing the value of rigorous evaluation beyond basic accuracy measures.

Activation Oracles (AOs) are promising methods for interpreting residual stream activations. However, current AOs face important issues, such as hallucinations and vagueness. Additionally, text-inversion confounds make them hard to evaluate. To this end, we improve the Activation Oracle (AO) training regime in four ways: training on on-policy rollouts, improving the conversational dataset, feeding more layers and an improvement to the injection formula. The capability improvements are marginal, but quality of life improvements are quite substantial. In addition, we open source the first comprehensive evaluation suite for AO quality, which we call AObench. Overall, we hope that our work sets a foundation that helps improve AOs and other models in the paradigm of scalable, end-to-end interpretability.

The robot learning community has made great strides in recent years, proposing new architectures and showcasing impressive new capabilities; however, the dominant metric used in the literature, especially for physical experiments, is "success rate", i.e. the percentage of runs that were successful. Furthermore, it is common for papers to report this number with little to no information regarding the number of runs, the initial conditions, and the success criteria, little to no narrative description of the behaviors and failures observed, and little to no statistical analysis of the findings. In this paper we argue that to move the field forward, researchers should provide a nuanced evaluation of their methods, especially when evaluating and comparing learned policies on physical robots. To do so, we propose best practices for future evaluations: explicitly reporting the experimental conditions, evaluating several metrics designed to complement success rate, conducting statistical analysis, and adding a qualitative description of failures modes. We illustrate these through an evaluation on physical robots of several learned policies for manipulation tasks.

Existing benchmarks measure capability -- whether a model succeeds on a single attempt -- but production deployments require reliability -- consistent success across repeated attempts on tasks of varying duration. We show these properties diverge systematically as task duration grows, and that pass@1 on short tasks is structurally blind to this divergence. We introduce a reliability science framework for long-horizon LLM agents with four metrics: Reliability Decay Curve (RDC), Variance Amplification Factor (VAF), Graceful Degradation Score (GDS), and Meltdown Onset Point (MOP). We evaluate 10 models across 23,392 episodes on a 396-task benchmark spanning four duration buckets and three domains. Key findings: (1) reliability decay is domain-stratified -- SE GDS drops from 0.90 to 0.44 while document processing is nearly flat (0.74 to 0.71); (2) VAF bifurcates by capability tier -- high VAF is a capability signature, not an instability signal; (3) capability and reliability rankings diverge substantially, with multi-rank inversions at long horizons; (4) frontier models have the highest meltdown rates (up to 19%) because they attempt ambitious multi-step strategies that sometimes spiral; and (5) memory scaffolds universally hurt long-horizon performance across all 10 models. These results motivate reliability as a first-class evaluation dimension alongside capability.

APR (Automated Program Repair) aims to automatically locate program defects, generate patches and validate the repairs. Existing techniques for APR are often combined with LLMs (Large Language Models), which leverages the code-related knowledge of LLMs to improve repair effectiveness. Current LLM-based APR methods typically utilize test cases only during the inference stage, adopting an iterative approach that performs repair first and validates it through test execution afterward. This conventional paradigm neglects two important aspects: the potential contribution of test cases in the training phase, and the possibility of leveraging testing prior to repair. To address this, we propose Repair-R1, which introduces test cases into the model's training phase and shifts test generation to precede repair. The model is required to first generate discriminative test cases that can distinguish defective behaviors, and then perform repair based on these tests. This enables the model to better locate defects and understand the underlying causes of defects, thereby improving repair effectiveness. We implement Repair-R1 with three different backbone models, using RL (reinforcement learning) to co-optimize test generation and bug repair. Experimental results on four widely adopted benchmarks demonstrate the superiority of Repair-R1. Specially, compared to vanilla models, Repair-R1 improves repair success rate by 2.68\% to 48.29\%, test generation success rate by 16.38\% to 53.28\%, and test coverage by 0.78\% to 53.96\%. We publish the code and weights at https://github.com/Tomsawyerhu/APR-RL and https://huggingface.co/tomhu/Qwen3-4B-RL-5000-step.

Large language model (LLM) agents are increasingly deployed to automate productivity tasks (e.g., email, scheduling, document management), but evaluating them on live services is risky due to potentially irreversible changes. Existing benchmarks rely on simplified environments and fail to capture realistic, stateful, multi-service workflows. We introduce ClawsBench, a benchmark for evaluating and improving LLM agents in realistic productivity settings. It includes five high-fidelity mock services (Gmail, Slack, Google Calendar, Google Docs, Google Drive) with full state management and deterministic snapshot/restore, along with 44 structured tasks covering single-service, cross-service, and safety-critical scenarios. We decompose agent scaffolding into two independent levers (domain skills that inject API knowledge via progressive disclosure, and a meta prompt that coordinates behavior across services) and vary both to measure their separate and combined effects. Experiments across 6 models, 4 agent harnesses, and 33 conditions show that with full scaffolding, agents achieve task success rates of 39-64% but exhibit unsafe action rates of 7-33%. On OpenClaw, the top five models fall within a 10 percentage-point band on task success (53-63%), with unsafe action rates from 7% to 23% and no consistent ordering between the two metrics. We identify eight recurring patterns of unsafe behavior, including multi-step sandbox escalation and silent contract modification.

Tool-using LLM agents can violate the very policies they are deployed to enforce while appearing to complete the task successfully. In policy-permissive environments, a tool may execute any well-formed call even when the corresponding state transition is forbidden by domain policy. The result is a silent wrong state (a booking cancelled, a passenger count changed, a claim acted on without verification) that neither the tool nor the agent's self-report exposes. We study this failure mode in the τ^2-bench airline domain. On a budget agent, 78% of observed failures are silent wrong-state failures with no tool error, and the aggregate failure rate is reproducible across disjoint seeds, not sampling noise. We then evaluate a lightweight intervention: deterministic, read-only pre-execution gates that inspect the proposed call and current state before allowing a write. A four-gate suite raises full-benchmark success from 29.6% to 42.0% on gpt-4o-mini (+12.4pp; paired task-level bootstrap P=0.0012), and the lift reproduces on a disjoint 15-seed set (+12.3pp; P=0.0008). The effect is concentrated where the gates fire: on the 26/50 firing tasks, success rises by +19.2pp, while movement on the 24 non-firing tasks does not exclude zero. Two negative controls (a self-enforcing retail domain and BFCL) bound the mechanism: gates help when tools are policy-permissive and add little where tools already self-enforce. As suggestive evidence, not a central claim, the same failure mode persists at the frontier: gpt-5.2 at default reasoning still attempts policy-violating writes, and the same suite improves success from 61.2% to 71.6% (+10.4pp; P=0.020; n=5, no replication). The contribution is a bounded evaluation and reliability result: deterministic gates do not guarantee task success, but they can deterministically prevent a known class of silent policy-violating writes at the action boundary.

We present and test the largest benchmark for vericoding, LLM-generation of formally verified code from formal specifications - in contrast to vibe coding, which generates potentially buggy code from a natural language description. Our benchmark contains 12,504 formal specifications, with 3,029 in Dafny, 2,334 in Verus/Rust and 7,141 in Lean. Of these, 6,174 are new unseen problems. We find vericoding success rates of 27% in Lean, 44% in Verus/Rust and 82% in Dafny using off-the-shelf LLMs. Adding natural-language descriptions does not significantly improve performance. We also find that LLM progress has improved progress on pure Dafny verification from 68% to 96% over the past year. The benchmark and vericoding results are shared at https://github.com/Beneficial-AI-Foundation/vericoding-benchmark

Large language models can perform well on many isolated tasks, yet they continue to struggle on multi-turn, long-horizon agentic problems that require skills such as planning, state tracking, and long context processing. In this work, we aim to better understand the relative importance of advancing these underlying capabilities for success on such tasks. We develop an oracle counterfactual framework for multi-turn problems that asks: how would an agent perform if it could leverage an oracle to perfectly perform a specific task? The change in the agent's performance due to this oracle assistance allows us to measure the criticality of such oracle skill in the future advancement of AI agents. We introduce a suite of procedurally generated, game-like tasks with tunable complexity. These controlled environments allow us to provide precise oracle interventions, such as perfect planning or flawless state tracking, and make it possible to isolate the contribution of each oracle without confounding effects present in real-world benchmarks. Our results show that while some interventions (e.g., planning) consistently improve performance across settings, the usefulness of other skills is dependent on the properties of the environment and language model. Our work sheds light on the challenges of multi-turn agentic environments to guide the future efforts in the development of AI agents and language models.

Large language models are moving from static text generators toward real-world decision-support systems, where forecasting is a composite capability that links information gathering, evidence integration, situational judgment, and action-oriented decision making. This capability is in broad demand across finance, policy, industry, and scientific research, yet its evaluation remains difficult: live benchmarks evaluate forecasts before answers exist, making them the cleanest way to measure forecasting ability, but they expire once events resolve; retrospective benchmarks are reproducible, but they cannot reliably distinguish genuine forecasting from facts a model may have already learned during pretraining. Prompting models to "pretend not to know" cannot replace a genuine knowledge boundary. We propose OracleProto, a reproducible framework for evaluating LLM native forecasting capability. OracleProto reconstructs resolved events into time-bounded forecasting samples by combining model-cutoff-aligned sample admission, tool-level temporal masking, content-level leakage detection, discrete answer normalization, and hierarchical scoring. Instantiated on a FutureX-Past-derived dataset with six contemporary LLMs, OracleProto distinguishes forecasting quality, sampling stability, and cost efficiency under controlled information boundaries, while reducing residual leakage to the 1% level, an order of magnitude below tool-only temporal filtering. OracleProto turns LLM forecasting from one-off evaluation into an auditable, reusable, and trainable dataset-level capability, providing a unified interface for fair cross-model comparison and a controlled signal source for downstream SFT and RL. Code and data are available at https://github.com/MaYiding/OracleProto and https://huggingface.co/datasets/MaYiding/OracleProto.

Large-scale AI training is now fundamentally a distributed systems problem, and hardware failures have become routine operating conditions rather than rare exceptions. Public operational evidence from production training clusters, however, remains scarce. This technical report presents an empirical analysis of a 63-node NVIDIA B200 production cluster (504 GPUs), using 55 days of Prometheus time-series data and 73 days of operational logs covering 224 multi-node training sessions. The cluster operates within a cross-organizational environment in which five parties (SKT, Upstage, Lablup, NVIDIA Korea, and VAST Data) share a unified monitoring pipeline. This arrangement enabled joint diagnosis of a 60-node-scale storage I/O bottleneck that did not appear at 2-4-node scale, a production-scale phenomenon no single team could isolate alone. Drawing on a months-long pre-training campaign, we perform three quantitative analyses yielding four findings. First, statistical analysis over 751 Prometheus metrics and 10 XID-identified GPU failures achieves a 10/10 detection rate (2/10 pre-XID) at ~0.84 false positives per day. No single metric is consistently dominant across failure types, motivating a multi-signal detection strategy. Second, profiling 523 checkpoint events along the GPU VRAM to NFS path attributes the "bandwidth paradox" (1.4-10.4% utilization of 200 Gbps RoCE) to saturation of the 128-slot NFS RPC layer. Third, multi-node failure response shows concentrated exclusions (top 3 of 63 nodes account for >50% of all exclusions) and an auto-retry chain success rate of 33.3% over 12 chains (73 attempts), 2.7x the 12.5% manual recovery rate; the median retry interval is 11 min (IQR 10-11). All analyses are grounded in production infrastructure providing session-level workload management, GPU-centric scheduling, and unified observability.

Developers increasingly delegate real maintenance work to product-grade coding agents, and many state tasks in their native language, in the style of a customer request rather than a curated English issue. We introduce RuBench 1.0, a benchmark of 25 tasks mined from recent fix commits in five live open-source repositories (aiohttp, aiogram, Laravel, NestJS, Fastify), each specified natively in Russian -- written from scratch, not translated -- and judged by the upstream maintainer's regression tests, which we withhold from release. All fix commits postdate the training-data cutoffs of every evaluated model. Round 1 evaluates Claude Code with Opus 4.8, Sonnet 5, and Haiku 4.5, and Codex CLI with GPT-5.5 (3 independent runs each; pass@1 with task-level uncertainty); the best configuration resolves 78.7% of tasks. Auditing full trajectories of an hors-concours configuration (Claude Code + Fable 5), we caught the product silently substituting the model on 20% of tasks via an official safeguard fallback -- evidence that the deployed product, not the model, is the unit actually measured. Version 2 adds Round 2: seven further configurations on the same frozen set under a per-configuration freshness gate -- the Russian-market agents SourceCraft CLI (ds, legacy) and Koda CLI (koda-pro), Antigravity with Gemini 3.1 Pro and 3.5 Flash, and Codex CLI with GPT-5.6 Sol and Luna. SourceCraft's flagship resolves 68.1% (N=23), above GPT-5.5 and both Gemini rows. A tool-call contamination re-audit of all 437 Round-2 trajectories finds the Russian and Gemini columns clean (0/293 cells) while flagging systematic oracle-hunting in the GPT-5.6 family (8/69 and 13/75 cells), including one case of mining a prior round's artifacts from the run machine's disk; honest scores are published alongside raw ones. We release statements, metadata, trajectories, and diffs; oracles are withheld with a SHA-256 manifest.

Agents based on Large Language Models (LLMs) have shown promise for performing sophisticated software engineering tasks autonomously. In addition, there has been progress towards developing agents that can perform parts of the research pipeline in machine learning and the natural sciences. We argue that research extension and its implementation is a critical capability for such systems, and introduce RExBench to support the evaluation of this capability. RExBench is a benchmark consisting of 12 realistic research experiment implementation tasks that aim to investigate research hypotheses that have not previously been implemented. Each task is set up as an extension to an existing research paper and codebase, accompanied by domain expert-written instructions. RExBench is robust to data contamination, and supports an automatic evaluation infrastructure that executes agent outputs to determine whether the success criteria are met. We use this benchmark to evaluate nine LLM agents implemented using three different frameworks: aider, Claude Code, and OpenHands. We find that all agents evaluated fail to autonomously implement the majority of the extensions. Although the success rate improves with additional human-written hints, the best performance under this setting remains below 40%. This indicates that current agents are still short of being able to handle realistic research extension tasks without substantial human guidance.

Large language models (LLMs) have advanced rapidly from conversational problem solving to addressing real-world tasks involving tool use, such as software engineering (SWE). Recent LLM-powered toolkits, such as OpenAI Codex and Cursor, have offered end-to-end automation of the software development process. However, building effective SWE agents remains challenging due to the lack of high-quality training data and effective test cases. To address this issue, we present SWE-Dev, an SWE agent built upon open-source LLMs. First, we develop a robust pipeline to synthesize test cases for patch evaluation. Second, we scale up agent trajectories to construct the training data for building SWE-Dev. Experiments on the SWE-bench-Verified benchmark show that the SWE-Dev models can achieve top performance among all open SWE agents. Specifically, the success rates of the SWE-Dev 7B and 32B parameter models reach 23.4% and 36.6%, respectively, outperforming state-of-the-art open-source models. All code, models, and datasets are publicly available at https://github.com/THUDM/SWE-Dev.

In modern Very Large Scale Integrated (VLSI) circuit design flow, the Register-Transfer Level (RTL) stage presents a critical opportunity for timing optimization. Addressing timing violations at this early stage is essential, as modern systems demand higher speeds, where even minor timing violations can lead to functional failures or system crashes. However, traditional timing optimization heavily relies on manual expertise, requiring engineers to iteratively analyze timing reports and debug. To automate this process, this paper proposes ViTAD, a method that efficiently analyzes the root causes of timing violations and dynamically generates targeted repair strategies. Specifically, we first parse Verilog code and timing reports to construct a Signal Timing Dependency Graph (STDG). Based on the STDG, we perform violation path analysis and use large language models (LLMs) to infer the root causes of violations. Finally, by analyzing the causes of violations, we selectively retrieve relevant debugging knowledge from a domain-specific knowledge base to generate customized repair solutions. To evaluate the effectiveness of our method, we construct a timing violation dataset based on real-world open-source projects. This dataset contains 54 cases of violations. Experimental results show that our method achieves a 73.68% success rate in repairing timing violations, while the baseline using only LLM is 54.38%. Our method improves the success rate by 19.30%.

Execution Accuracy (EX), the widely used metric for evaluating the effectiveness of Natural Language to SQL (NL2SQL) solutions, is becoming increasingly unreliable. It is sensitive to syntactic variation, ignores that questions may admit multiple interpretations, and is easily misled by erroneous ground-truth SQL. To address this, we introduce ROSE, an intent-centered metric that focuses on whether the predicted SQL answers the question, rather than consistency with the ground-truth SQL under the reference-dependent paradigm. ROSE employs an adversarial Prover-Refuter cascade: SQL Prover assesses the semantic correctness of a predicted SQL against the user's intent independently, while Adversarial Refuter uses the ground-truth SQL as evidence to challenge and refine this judgment. On our expert-aligned validation set ROSE-VEC, ROSE achieves the best agreement with human experts, outperforming the next-best metric by nearly 24% in Cohen's Kappa. We also conduct a largescale re-evaluation of 19 NL2SQL methods, revealing four valuable insights. We release ROSE and ROSE-VEC to facilitate more reliable NL2SQL research.

This work establishes that trigger-word data poisoning of vision language action models is practical, while at the same time the open-source robotics ecosystem holds trust assumptions about community contributions. A few poisoned samples can silently embed a backdoor that disables a robot on command. We evaluate this threat against smolVLA on a real-world pick-and-place task, training on three poison ratios and evaluating across different prompts on the LeRobot platform. Three poisoned episodes in 320 clean episodes suffice for a complete denial of service. Success rate drops to 0.0 plus minus 0.0% across all trigger-word conditions and the robot locks into a fixed joint configuration rather than executing any task-relevant motion. Clean-prompt behaviour holds at approx. 50% success rate across all poison ratios, confirming the attack is stealthy under normal operation. A single poisoned episode already reduces success rate to 6.7 plus minus 6.7%. The robot still moves, but no longer completes the task. The attack generalises to front, middle, and end trigger placements despite training exclusively on front-placed triggers. These findings establish that the threat is practical, low-cost, and stealthy, and warrant treating dataset provenance as a first-class concern in open-source robotics ecosystems.

Enterprise ERP systems managing hundreds of thousands of employee records face critical data quality challenges when human resources departments perform decentralized manual entry across multiple languages. We present an end-to-end pipeline combining automated data cleaning with LLM-driven SQL query generation, deployed on a production system managing 240,000 employee records over six months. The system operates in two integrated stages: a multi-stage cleaning pipeline that performs translation normalization, spelling correction, and entity deduplication during periodic synchronization from Microsoft SQL Server to PostgreSQL; and a retrieval-augmented generation framework powered by GPT-4o that translates natural-language questions in Turkish, Russian, and English into validated SQL queries. The query engine employs LangChain orchestration, FAISS vector similarity search, and few-shot learning with 500+ validated examples. Our evaluation demonstrates 92.5% query validity, 95.1% schema compliance, and 90.7\% semantic accuracy on 2,847 production queries. The system reduces query turnaround time from 2.3 days to under 5 seconds while maintaining 99.2% uptime, with GPT-4o achieving 46% lower latency and 68% cost reduction versus GPT-3.5. This modular architecture provides a reproducible framework for AI-native enterprise data governance, demonstrating real-world viability at enterprise scale with 4.3/5.0 user satisfaction.

Real-world clinical text-to-SQL requires reasoning over heterogeneous EHR tables, temporal windows, and patient-similarity cohorts to produce executable queries. We introduce CLINSQL, a benchmark of 633 expert-annotated tasks on MIMIC-IV v3.1 that demands multi-table joins, clinically meaningful filters, and executable SQL. Solving CLINSQL entails navigating schema metadata and clinical coding systems, handling long contexts, and composing multi-step queries beyond traditional text-to-SQL. We evaluate 22 proprietary and open-source models under Chain-of-Thought self-refinement and use rubric-based SQL analysis with execution checks that prioritize critical clinical requirements. Despite recent advances, performance remains far from clinical reliability: on the test set, GPT-5-mini attains 74.7% execution score, DeepSeek-R1 leads open-source at 69.2% and Gemini-2.5-Pro drops from 85.5% on Easy to 67.2% on Hard. Progress on CLINSQL marks tangible advances toward clinically reliable text-to-SQL for real-world EHR analytics.

Text-to-SQL technology has become crucial for translating natural language into SQL queries in various industries, enabling non-technical users to perform complex data operations. The need for accurate evaluation methods has increased as these systems have grown more sophisticated. However, we found that the Execution Accuracy (EX), the most promising evaluation metric, still shows a substantial portion of false positives and negatives compared to human evaluation. Thus, this paper introduces FLEX (False-Less EXecution), a novel approach to evaluating text-to-SQL systems using large language models (LLMs) to emulate human expert-level evaluation of SQL queries. Our method shows significantly higher agreement with human expert judgments, improving Cohen's kappa from 61 to 78.17. Re-evaluating top-performing models on the Spider and BIRD benchmarks using FLEX reveals substantial shifts in performance rankings, with an average performance decrease of 3.15 due to false positive corrections and an increase of 6.07 from addressing false negatives. This work contributes to a more accurate and nuanced evaluation of text-to-SQL systems, potentially reshaping our understanding of state-of-the-art performance in this field.

Wu et al. (2026) showed that most frontier large language models (LLMs) recommend a sponsored, roughly twice-as-expensive flight when their system prompt contains a soft sponsorship cue. We reproduce their evaluation on ten open-weight chat models plus the two of their twenty-three models that are still reachable today (gpt-3.5-turbo, gpt-4o). All reported rates in this paper are produced under the same judge the original paper used (gpt-4o); we additionally store every label under an open-weight (gpt-oss-120b) and a smaller proprietary (gpt-4o-mini) judge for an ablation. Three findings emerge. First, a prose description of an LLM evaluation pipeline is not, on its own, sufficient for accurate reproduction: we surfaced three silent implementation failures that each shifted a reported rate by tens of percentage points. Second, the central claims do generalise - the gpt-3.5-turbo logistic-regression intercept of alpha = 0.81 is within four points of the original alpha = 0.86, and 200 of 200 trials on gpt-3.5-turbo and gpt-4o promote a payday lender to a financially distressed user. Third, a thirty-token user prompt that asks the assistant for a neutral comparison table first cuts sponsored recommendation from 46.9% to 1.0% averaged across our ten open-source models, and from 53.0% to 0% averaged across the two OpenAI models. AI literacy and price-comparison portals are likely market-level mitigations; the harmful-product cell is bounded by neither. Raw data, labels and analysis scripts are at https://github.com/akmaier/Paper-LLM-Ads .

Machine learning has demonstrated remarkable performance over finite datasets, yet whether the scores over the fixed benchmarks can sufficiently indicate the model's performance in the real world is still in discussion. In reality, an ideal robust model will probably behave similarly to the oracle (e.g., the human users), thus a good evaluation protocol is probably to evaluate the models' behaviors in comparison to the oracle. In this paper, we introduce a new robustness measurement that directly measures the image classification model's performance compared with a surrogate oracle (i.e., a foundation model). Besides, we design a simple method that can accomplish the evaluation beyond the scope of the benchmarks. Our method extends the image datasets with new samples that are sufficiently perturbed to be distinct from the ones in the original sets, but are still bounded within the same image-label structure the original test image represents, constrained by a foundation model pretrained with a large amount of samples. As a result, our new method will offer us a new way to evaluate the models' robustness performance, free of limitations of fixed benchmarks or constrained perturbations, although scoped by the power of the oracle. In addition to the evaluation results, we also leverage our generated data to understand the behaviors of the model and our new evaluation strategies.

Existing benchmarks for tool-using LLM agents primarily report single-run success rates and miss reliability properties required in production. We introduce ReliabilityBench, a benchmark for evaluating agent reliability across three dimensions: (i) consistency under repeated execution using pass^k, (ii) robustness to semantically equivalent task perturbations at intensity ε, and (iii) fault tolerance under controlled tool/API failures at intensity λ. ReliabilityBench contributes a unified reliability surface R(k,ε,λ), action metamorphic relations that define correctness via end-state equivalence rather than text similarity, and a chaos-engineering-style fault injection framework (timeouts, rate limits, partial responses, schema drift). We evaluate two models (Gemini 2.0 Flash, GPT-4o) and two agent architectures (ReAct, Reflexion) across four domains (scheduling, travel, customer support, e-commerce) over 1,280 episodes. Perturbations alone reduce success from 96.9% at ε=0 to 88.1% at ε=0.2. Rate limiting is the most damaging fault in ablations. ReAct is more robust than Reflexion under combined stress, and Gemini 2.0 Flash achieves comparable reliability to GPT-4o at much lower cost. ReliabilityBench provides a systematic framework for assessing production readiness of LLM agents.

Text-to-SQL task maps natural language utterances to structured queries that can be issued to a database. State-of-the-art (SOTA) systems rely on finetuning large, pre-trained language models in conjunction with constrained decoding applying a SQL parser. On the well established Spider dataset, we begin with Oracle studies: specifically, choosing an Oracle hypothesis from a SOTA model's 10-best list, yields a 7.7% absolute improvement in both exact match (EM) and execution (EX) accuracy, showing significant potential improvements with reranking. Identifying coherence and correctness as reranking approaches, we design a model generating a query plan and propose a heuristic schema linking algorithm. Combining both approaches, with T5-Large, we obtain a consistent 1% improvement in EM accuracy, and a ~2.5% improvement in EX, establishing a new SOTA for this task. Our comprehensive error studies on DEV data show the underlying difficulty in making progress on this task.

Researchers have proposed numerous text-to-SQL techniques to streamline data analytics and accelerate the development of data-driven applications. To compare these techniques and select the best one for deployment, the community depends on public benchmarks and their leaderboards. Since these benchmarks heavily rely on human annotations during question construction and answer evaluation, the validity of the annotations is crucial. In this paper, we conduct an empirical study that (i) benchmarks annotation error rates for two widely used text-to-SQL benchmarks, BIRD and Spider 2.0-Snow, and (ii) corrects a subset of the BIRD development (Dev) set to measure the impact of annotation errors on text-to-SQL agent performance and leaderboard rankings. Through expert analysis, we show that BIRD Mini-Dev and Spider 2.0-Snow have error rates of 52.8% and 62.8%, respectively. We re-evaluate all 16 open-source agents from the BIRD leaderboard on both the original and the corrected BIRD Dev subsets. We show that performance changes range from -7% to 31% (in relative terms) and rank changes range from -9 to +9 positions. We further assess whether these impacts generalize to the full BIRD Dev set. We find that the rankings of agents on the uncorrected subset correlate strongly with those on the full Dev set (Spearman's r_s=0.85, p=3.26e-5), whereas they correlate weakly with those on the corrected subset (Spearman's r_s=0.32, p=0.23). These findings show that annotation errors can significantly distort reported performance and rankings, potentially misguiding research directions or deployment choices. Our code and data are available at https://github.com/uiuc-kang-lab/text_to_sql_benchmarks.

Java remains central to enterprise software, and many applications outlive their original architecture. Migrating them across frameworks is a behavior-preserving refactoring spanning build configuration, dependency injection, persistence, request handling, and deployment. Existing software-engineering benchmarks cover bug fixing, feature implementation, and language or version modernization, but leave cross-framework refactoring largely unmeasured. We introduce ScarfBench, a benchmark for behavior-preserving cross-framework refactoring of enterprise Java applications. It is built from expert-written implementation triples across Spring, Jakarta EE, and Quarkus: 34 applications (29 focused single-layer, 5 whole) yielding 102 variants (~151K lines across 1946 source and test files) and 204 directed refactoring tasks. Each task gives an agent a working source application and a target framework; the agent must synthesize a target implementation preserving the source behavior. Correctness is evaluated by an application-specific executable oracle: the candidate must compile, deploy in a containerized target runtime, and pass behavioral tests over the application's observable interface. We evaluate five state-of-the-art coding agents on ScarfBench. The strongest achieves only 15.3% aggregate test pass on focused-layer migrations and 12.2% on whole applications, and only one of the 204 tasks yields a fully behaviorally equivalent target. Difficulty is asymmetric across framework directions and architectural layers: Spring<->Quarkus is the most tractable pair, and Jakarta-targeted migrations are hardest. From LLM-as-a-judge and expert adjudication of failed-task traces, we derive a taxonomy of recurring failure categories spanning build, deploy, and test stages. We release the benchmark, harness, and agent traces at https://scarfbench.info.

Personal AI agents like OpenClaw run with elevated privileges on users' local machines, where a single successful prompt injection can leak credentials, redirect financial transactions, or destroy files. This threat goes well beyond conventional text-level jailbreaks, yet existing safety evaluations fall short: most test models in isolated chat settings, rely on synthetic environments, and do not account for how the agent framework itself shapes safety outcomes. We introduce CLAWSAFETY, a benchmark of 120 adversarial test scenarios organized along three dimensions (harm domain, attack vector, and harmful action type) and grounded in realistic, high-privilege professional workspaces spanning software engineering, finance, healthcare, law, and DevOps. Each test case embeds adversarial content in one of three channels the agent encounters during normal work: workspace skill files, emails from trusted senders, and web pages. We evaluate five frontier LLMs as agent backbones, running 2,520 sandboxed trials across all configurations. Attack success rates (ASR) range from 40\% to 75\% across models and vary sharply by injection vector, with skill instructions (highest trust) consistently more dangerous than email or web content. Action-trace analysis reveals that the strongest model maintains hard boundaries against credential forwarding and destructive actions, while weaker models permit both. Cross-scaffold experiments on three agent frameworks further demonstrate that safety is not determined by the backbone model alone but depends on the full deployment stack, calling for safety evaluation that treats model and framework as joint variables. Code and data will be available at: https://weibowen555.github.io/ClawSafety/.

Large language models (LLMs) are increasingly used for tasks that require complex reasoning. Most benchmarks focus on final outcomes but overlook the intermediate reasoning steps - such as planning, revision, and decision making under resource constraints. We argue that measuring these internal processes is essential for understanding model behavior and improving reliability. We propose using strategic games as a natural evaluation environment: closed, rule-based systems with clear states, limited resources, and automatic feedback. We introduce a framework that evaluates LLMs along three core dimensions: planning, revision, and resource-constrained decision making. To operationalize this, we define metrics beyond win rate, including overcorrection risk rate, correction success rate, improvement slope, and over-budget ratio. In 4320 adversarial rounds across 12 leading models, ChatGPT-o3-mini achieves the top composite score, with a win rate of 74.7 percent, a correction success rate of 78.6 percent, and an improvement slope of 0.041. By contrast, Qwen-Plus, despite an overcorrection risk rate of 81.6 percent, wins only 25.6 percent of its matches - primarily due to excessive resource use. We also observe a negative correlation between overcorrection risk rate and correction success rate (Pearson r = -0.51, p = 0.093), suggesting that more frequent edits do not always improve outcomes. Our findings highlight the value of assessing not only what LLMs decide but how they arrive at those decisions