paper_id stringlengths 10 21 | title stringlengths 10 177 | paper_url stringlengths 42 53 | authors listlengths 1 51 | type stringclasses 3
values | primary_area stringclasses 0
values | abstract large_stringlengths 358 2.69k | keywords listlengths 0 0 | TL;DR large_stringclasses 0
values | submission_number int64 -35,412 36.8k ⌀ | arxiv_id stringlengths 10 10 ⌀ | arxiv_id_source stringclasses 1
value |
|---|---|---|---|---|---|---|---|---|---|---|---|
sHRb3N6tLL | Pixel-space Autoregressive Image Synthesis via Spectrum Serialization and Flow-based Refinement | https://openreview.net/forum?id=sHRb3N6tLL | [
"Guiwei Zhang",
"Tianyu Zhang",
"Yalong Bai",
"Ying Ba",
"Zichang Tan",
"Yang Yang"
] | Poster | null | Autoregressive (AR) image generators commonly rely on vector-quantized (VQ) autoencoders to compress images into discrete token sequences. However, quantization inevitably discards visual information, and the resulting reconstruction errors may propagate through AR decoding, limiting generation fidelity. We revisit the... | [] | null | 15,017 | null | null |
LAS8pm4WJ6 | Do Latent-CoT Models Think Step-by-Step? A Mechanistic Study on Sequential Reasoning Tasks | https://openreview.net/forum?id=LAS8pm4WJ6 | [
"Jia Liang",
"Liangming Pan"
] | Poster | null | Latent Chain-of-Thought aims to enable step-by-step computation without emitting long rationales, yet its internal mechanisms remain unclear. We study CODI, a continuous-thought teacher–student distillation model, on strictly sequential polynomial-iteration tasks with known intermediate states. Using logit-lens decodin... | [] | null | 20,113 | 2602.00449 | title_snapshot |
4ElMzrtekX | Compositional Reasoning in Language Models under Reinforcement Learning Post-Training | https://openreview.net/forum?id=4ElMzrtekX | [
"Yu He",
"Yingxi Li",
"Yifei Wang",
"Ellen Vitercik"
] | Poster | null | Compositional reasoning is critical for real-world problem solving: since training data is necessarily limited, models must generalize by composing learned skills in new ways. While post-training methods such as reinforcement learning (RL) have substantially improved the reasoning abilities of language models (LMs), th... | [] | null | 23,299 | 2609.19465 | title_snapshot |
QG31f899dE | Minimax Optimal Estimation of Transport-Growth Pairs in Unbalanced Optimal Transport | https://openreview.net/forum?id=QG31f899dE | [
"Donlapark Ponnoprat",
"Noboru Isobe",
"Masaaki Imaizumi"
] | Poster | null | Unbalanced optimal transport extends classical optimal transport to measures with different total masses, but statistical guarantees for Monge-type estimation remain limited. We study unbalanced transport with quadratic cost and Kullback-Leibler marginal penalties and argue that the natural population target is not a m... | [] | null | 6,463 | 2605.08705 | title_snapshot |
ybJgoCc3bI | UniFlowDock: Flexible Docking with Complete Equivariant Velocity Fields | https://openreview.net/forum?id=ybJgoCc3bI | [
"Kangxin Chen",
"Jieyu Zhao",
"Xulun Ye",
"Kun Zhou",
"Jinli Hu",
"Yuanyuan Deng",
"Min Xie"
] | Poster | null | Flexible molecular docking is crucial for drug discovery; however, existing flow-matching generative models often produce physically invalid conformations (e.g., steric clashes, distorted bond geometry). Our analysis suggests that a primary contributor to this challenge lies at the modeling level: when the flow is para... | [] | null | 13,219 | null | null |
OEIZ6QQAsu | Optimal In-Context Learning of Autoregressive Processes under Heterogeneous Second-Order Moments of the Prompts | https://openreview.net/forum?id=OEIZ6QQAsu | [
"Hanna Tseran",
"Masaaki Imaizumi"
] | Poster | null | In-context learning enables large language models to adapt to tasks directly from input sequences, without parameter updates. We investigate the mechanism underlying in-context learning of autoregressive processes in transformers under heterogeneous second-order moments of prompt distributions. Our study considers a tw... | [] | null | 10,555 | null | null |
C8agNRzT7b | Learning Compositional Latent Structure with Vector Networks | https://openreview.net/forum?id=C8agNRzT7b | [
"Niclas Pokel",
"Benjamin F. Grewe"
] | Poster | null | Deep networks are powerful function approximators, but they typically store many different computations in shared weight matrices, making it difficult to selectively reuse or adapt parts of them when a familiar structure appears in novel combinations. We introduce the Vector Network (VN), a hierarchical recurrent archi... | [] | null | 14,684 | 2605.28007 | title_snapshot |
jIJ566xVDK | Adaptive Delayed-Update Cyclic Algorithm for Variational Inequalities | https://openreview.net/forum?id=jIJ566xVDK | [
"Yi Wei",
"Xufeng Cai",
"Jelena Diakonikolas"
] | Poster | null | Cyclic block coordinate methods are widely used in practice for their simplicity and strong empirical performance. Yet, their theoretical behavior is challenging to explain, and setting their step sizes$-$beyond classical coordinate descent for minimization$-$requires careful tuning or line-search machinery. In this wo... | [] | null | 7,218 | 2603.29128 | title_snapshot |
f21HhDeBNh | Axiomatic World Modeling for Physics Reasoning | https://openreview.net/forum?id=f21HhDeBNh | [
"Xinye Yang",
"Zhenyang Liu",
"Yuxuan Wang",
"Yuanyuan Lei"
] | Poster | null | Recent research-level physics-reasoning benchmarks for LLMs reveal a failure mode that lies not in knowledge or execution, but in modeling the world being reasoned about. When faced with unseen problem setups, models often silently alter given quantities, fabricate laws, or inject unstated assumptions to force the prob... | [] | null | 30,863 | null | null |
wfY4Srt4qX | Opening the Black Box of Classifier-Free Guidance via Information Bottleneck | https://openreview.net/forum?id=wfY4Srt4qX | [
"Jiayang Gao",
"Tianyi Zheng",
"Jiayang Zou",
"Fengxiang Yang",
"Luyao Fan",
"Lv Tang",
"Bo Li",
"Jia Wang"
] | Poster | null | Classifier-free guidance (CFG) is a standard technique for improving conditional generation in diffusion models, yet its activation schedule is usually selected heuristically. In practice, most existing CFG-based methods rely on a pre-defined activation policy throughout inference, while recent dynamic variants often d... | [] | null | 1,098 | 2610.13369 | title_snapshot |
nlaQh30azU | GRNAgent: A Multimodal Graph Reasoning Agent for Gene Regulatory Network Inference | https://openreview.net/forum?id=nlaQh30azU | [
"Akshata Hegde",
"Kyler Zook",
"Yanli Wang",
"Jianlin Cheng"
] | Poster | null | Inferring gene regulatory networks (GRNs) from multimodal biological data requires integrating regulatory evidence beyond transcript abundance alone. Regulatory evidence can come from RNA-seq, chromatin accessibility, and transcription factor (TF) motif priors, but these sources are difficult to discover, harmonize, an... | [] | null | 30,536 | null | null |
j7Y9kxJJPM | SIEVES: Selective Prediction Generalizes through Visual Evidence Scoring | https://openreview.net/forum?id=j7Y9kxJJPM | [
"Hector G. Rodriguez",
"Marcus Rohrbach"
] | Poster | null | Multimodal large language models (MLLMs) achieve ever-stronger performance on visual-language tasks. Even as traditional visual question answering benchmarks approach saturation, reliable deployment requires satisfying low error tolerances in real-world out-of-distribution (OOD) scenarios. Precisely, selective predicti... | [] | null | 24,753 | 2604.25855 | title_snapshot |
TqEugPEDx7 | Fine-Tuning Improves Information Conveyance in Language Models | https://openreview.net/forum?id=TqEugPEDx7 | [
"Yuwei Cheng",
"Weiyi Tian",
"Haifeng Xu"
] | Poster | null | Fine-tuning is often believed to reduce uncertainty and diversity in large language models, but existing analyses overlook output length, a key confounder, and therefore fail to capture how uncertainty is distributed across an entire generation rollout. To address this, we propose Canopy Entropy ($\mathrm{CE}^\star$), ... | [] | null | 7,364 | 2605.30844 | title_snapshot |
up4CskJQ0w | MMSkills: Towards Multimodal Skills for General Visual Agents | https://openreview.net/forum?id=up4CskJQ0w | [
"Kangning Zhang",
"Shuai Shao",
"Wenxiang Jiao",
"Qingyao Li",
"Jianghao Lin",
"Lingyue Fu",
"Shijian Wang",
"Yuan Lu",
"Weiwen Liu",
"Weinan Zhang",
"Yong Yu"
] | Poster | null | Reusable skills have become a core substrate for improving agent capabilities, yet most existing skill packages encode reusable behavior primarily as textual prompts, executable code, or learned routines. For visual agents, however, procedural knowledge is inherently multimodal: reuse depends not only on what operation... | [] | null | 2,123 | 2605.13527 | title_snapshot |
5idvmMr6r1 | OASIS: Online Adaptive Steering for In-Training Safety of LLMs | https://openreview.net/forum?id=5idvmMr6r1 | [
"Yifan Sun",
"Qiang Sheng",
"Ya Wu",
"Zhengjia Wang",
"Guang Yang",
"Shaofei Wang",
"Danding Wang",
"Juan Cao"
] | Poster | null | Fine-tuning is essential for adapting Large Language Models to downstream tasks. However, this process can also inadvertently erode the critical safety alignment even when the fine-tuning data appears benign. Prior methods either introduce safety regularizers that may conflict with the primary training objective, or re... | [] | null | 535 | null | null |
fg8I6pwObp | Robust Policy Optimization to Prevent Catastrophic Forgetting | https://openreview.net/forum?id=fg8I6pwObp | [
"Mahdi Sabbaghi",
"George J. Pappas",
"Adel Javanmard",
"Hamed Hassani"
] | Poster | null | LLMs are commonly trained through multi-stage post-training: first via RLHF, then fine-tuned for other downstream objectives. Yet even small downstream updates can compromise earlier learned behaviors (e.g., safety), exposing a brittleness known as catastrophic forgetting. This suggests standard RLHF objectives do not ... | [] | null | 29,580 | 2602.08813 | title_snapshot |
4UGod3zYeD | InfoSFT: Learn More and Forget Less with Information-Aware Token Weighting | https://openreview.net/forum?id=4UGod3zYeD | [
"Mahdi Sabbaghi",
"George J. Pappas",
"Adel Javanmard",
"Hamed Hassani"
] | Poster | null | Supervised fine-tuning (SFT) provides the standard approach for teaching LLMs new behaviors from offline expert demonstrations. However, standard SFT uniformly fits all samples—including those with low likelihood under the base model—which can disproportionately drive training updates toward overfitting specific sample... | [] | null | 30,503 | 2605.14967 | title_snapshot |
T0OsmO746x | DiDE: Direct Injection with Color-Texture DEcoupling for 3D Stylization | https://openreview.net/forum?id=T0OsmO746x | [
"Tao Wu",
"Alexandra Gomez-Villa",
"Senmao Li",
"Yaxing Wang",
"Joost van de Weijer",
"Kai Wang"
] | Poster | null | Recent advances in rectified flow-based image-to-3D generative models have enabled high-fidelity 3D asset generation. Building on this, a growing line of work has exploited these strong 3D priors for training-free stylization, transferring visual attributes from a reference image onto a generated 3D asset. However, exi... | [] | null | 1,305 | 2610.02044 | title_snapshot |
x0pAg7jich | Jacobian Scopes: A Unified Geometric Framework for Token-Level LLM Attributions | https://openreview.net/forum?id=x0pAg7jich | [
"Toni Liu",
"Baran Zadeoğlu",
"Nicolas Boulle",
"Raphaël Sarfati",
"Gurbir Arora",
"Christopher Earls"
] | Poster | null | Gradient-based attribution methods for large language models (LLMs) have so far been limited to explaining scalar outputs - the logit of a single target token. Yet LLM predictions are inherently distributional, and many practically important questions concern the full predictive distribution: which tokens make the mode... | [] | null | 30,235 | null | null |
UmuBorIdCx | Co-Evolving Policy Distillation | https://openreview.net/forum?id=UmuBorIdCx | [
"Naibin Gu",
"Chenxu Yang",
"Qingyi Si",
"Chuanyu Qin",
"Dingyu Yao",
"Peng Fu",
"Zheng Lin",
"Weiping Wang",
"Nan Duan",
"Jiaqi Wang"
] | Poster | null | RLVR and OPD have become standard paradigms for post-training. We provide a unified analysis of these two paradigms in consolidating multiple expert capabilities into a single model, identifying capability loss in different ways: mixed RLVR suffers from inter-capability divergence cost, while the pipeline of first trai... | [] | null | 20,208 | 2604.27083 | title_snapshot |
TAt6jrbMb8 | Self-Distilled RLVR | https://openreview.net/forum?id=TAt6jrbMb8 | [
"Chenxu Yang",
"Chuanyu Qin",
"Qingyi Si",
"Minghui Chen",
"Naibin Gu",
"Dingyu Yao",
"Zheng Lin",
"Weiping Wang",
"Nan Duan",
"Jiaqi Wang"
] | Spotlight | null | On-policy distillation (OPD) has become a popular training paradigm in the LLM community. This paradigm selects a larger model as the teacher to provide dense, fine-grained signals for each sampled trajectory, in contrast to reinforcement learning with verifiable rewards (RLVR), which only obtains sparse signals from v... | [] | null | 21,434 | 2604.03128 | title_snapshot |
kYO08R5mas | Learning Hierarchical Forward Processes For Discrete Diffusion Language Models | https://openreview.net/forum?id=kYO08R5mas | [
"Inhyeok Jeong",
"Jeongwhan Choi",
"Jaehyeon Park",
"Noseong Park"
] | Poster | null | Discrete diffusion language models (DLMs) enable parallel generation but usually rely on fixed forward processes, such as masking, uniform noise, or precomputed hierarchies. We investigate whether the hierarchical forward process can instead be learned while preserving continuous-time tractability. We propose \textbf{L... | [] | null | 24,577 | null | null |
XHd7CqQtoF | CoScan: Multi-Scale Content-Adaptive Space-Filling Scans for Causal State-Space Image Restoration | https://openreview.net/forum?id=XHd7CqQtoF | [
"Wei Dong",
"Terry Ji",
"Yan Min",
"Shahram Shirani",
"Jun Chen",
"Han Zhou"
] | Poster | null | Selective state-space models (SSMs) have recently shown strong potential for efficient image restoration. A central difficulty, however, is that a 2D feature map typically needs to be linearized into a 1D token sequence for causal state-space updates, making restoration performance highly sensitive to the imposed scan ... | [] | null | 809 | null | null |
lpI5QYN2jv | Think, Plan, Paint: Layout-Aware Reasoning for Controllable Image Generation in Unified Models | https://openreview.net/forum?id=lpI5QYN2jv | [
"Junhao Liu",
"Jian-Wei Zhang",
"Tao Huang",
"Miles Yang",
"Zhao Zhong",
"Liefeng Bo"
] | Poster | null | Unified Multimodal Large Language Models (MLLMs) offer a promising paradigm for unifying visual understanding and generation, yet they still struggle to follow complex spatial instructions and logical constraints in controllable image generation. To address this gap, we present ATLAS, a unified framework that equips ML... | [] | null | 1,509 | 2607.16409 | title_snapshot |
eu0NxAFRN2 | Improving Context-Shift Robustness of Convolutional Networks via Context-Regularized Cross-Entropy | https://openreview.net/forum?id=eu0NxAFRN2 | [
"Jinen Setpal",
"Chaoyue Liu"
] | Poster | null | Image classification models often encounter objects in diverse and unpredictable contexts at test time, which may differ substantially from those seen during training. This phenomenon, termed context shift, exposes a key fragility in neural networks: when models rely on non-causal contextual cues for prediction, shifts... | [] | null | 12,679 | null | null |
g7cJBv2nOf | Geometry over Density: Few-Shot Cross-Domain OOD Detection | https://openreview.net/forum?id=g7cJBv2nOf | [
"Li Li",
"You Qin",
"Jiate Li",
"Charith Peris",
"Lisa Bauer",
"Roger Zimmermann",
"Yue Zhao"
] | Poster | null | Out-of-distribution (OOD) detection identifies test samples that fall outside a model's training distribution, a capability critical for safe deployment in high-stakes applications. Standard OOD detectors are trained on a specific in-distribution (ID) dataset and detect deviations from that single domain. In contrast, ... | [] | null | 3,324 | 2605.03410 | title_snapshot |
tgSNcsvbz9 | Learning to play with spikes. Characterizing, predicting, and engineering unsupervised plasticity rules for spiking reservoir computing | https://openreview.net/forum?id=tgSNcsvbz9 | [
"Maciej Kania",
"Basile Confavreux",
"Tim Vogels"
] | Poster | null | Spiking reservoir computing, and reservoir computing more generally, is a powerful and efficient framework for neuromorphic and biological applications, in which a fixed random reservoir drives a trained readout. Yet its performance depends critically on the reservoir initialization. Enriching the reservoir with adapti... | [] | null | 34,311 | null | null |
t5syQ7vPjU | All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow | https://openreview.net/forum?id=t5syQ7vPjU | [
"Wei Dong",
"Han Zhou",
"Terry Ji",
"Guanhua Zhao",
"Shahab Asoodeh",
"Yulun Zhang",
"Guangtao Zhai",
"Jun Chen",
"Xiaohong Liu"
] | Poster | null | Adverse weather removal (AWR) in real-world images remains challenging due to heterogeneous and unseen degradations, while distortion-driven training often yields overly smooth results. We propose PVRF, a unified framework that integrates zero-shot soft weather Perceptions with Velocity-constrained Rectified-Flow refin... | [] | null | 779 | null | null |
vP6mCzbls0 | Cross-Foundation Complementary Learning Systems for Continual Test-Time Adaptation in Open-Vocabulary Semantic Segmentation | https://openreview.net/forum?id=vP6mCzbls0 | [
"Guan-Yu Chen",
"Wei-Hua Li",
"Huei-Fang Yang",
"Chu-Song Chen"
] | Poster | null | Open-vocabulary semantic segmentation (OVSS) models enable dense prediction for classes specified at test time, but continual test-time adaptation (CTTA) over long, non-stationary streams can corrupt their language-aligned semantic interface, which we diagnose through synonym-prompt text alignment. We propose Cross-Fou... | [] | null | 26,268 | null | null |
eKJJOkZmQA | An Analytical Model of Compute-limited Multistage Training Pipelines | https://openreview.net/forum?id=eKJJOkZmQA | [
"Nishil Patel",
"Jin Hwa Lee",
"Basile Confavreux",
"Andrew Saxe"
] | Poster | null | Modern foundation models are trained through multistage pipelines, typically beginning with unsupervised pretraining, followed by supervised fine-tuning (SFT) on step-by-step solutions, and reinforcement learning (RL) with bulk feedback. Empirically, the design of these stages, particularly the compute allocated to eac... | [] | null | 21,713 | null | null |
Ko8MeBmiba | Towards Identifying Dominant Low-Rank Subspaces in Zeroth-Order Fine-Tuning | https://openreview.net/forum?id=Ko8MeBmiba | [
"Jinjie Fang",
"Chengxun Jin",
"Yi Chang",
"Bin Gu"
] | Poster | null | Adapting Large Language Models (LLMs) to downstream tasks is increasingly bottlenecked by the staggering memory overhead of first-order (FO) backpropagation. Zeroth-order (ZO) optimization has emerged as a compelling, memory-efficient alternative; however, it suffers from the curse of dimensionality, which introduces p... | [] | null | 15,895 | null | null |
dMShldodM4 | Unifying Sparsity and Discreteness: One-Shot Pruning for Quantized LLMs via Discrete Optimization | https://openreview.net/forum?id=dMShldodM4 | [
"Haozhen Zhang",
"Hanyuan Zheng",
"Teng Hou",
"Zhaogeng Liu",
"Yi Chang",
"Bin Gu"
] | Poster | null | Pruning and quantization are two dominant techniques that effectively address the computational and storage burdens of large language model (LLM) inference on edge devices. Recently, one-shot pruning has gained particular attention for its ability to identify weight supports via optimization without retraining. However... | [] | null | 15,877 | null | null |
PxiWypLBRI | GADMVP: Adaptive Few-Shot Graph-Level Anomaly Detection with Multi-View Structured Prompting | https://openreview.net/forum?id=PxiWypLBRI | [
"Xiaolin Han",
"Xiurui Hu",
"Lingyun Song",
"Yudai Pan",
"Xuequn Shang"
] | Poster | null | Graph-level anomaly detection plays a critical role in a wide range of applications, such as fraud detection in financial networks and malicious program detection, yet it remains challenging in real-world scenarios where labeled anomalous graphs are extremely scarce and graph structures are highly diverse. Existing few... | [] | null | 31,888 | null | null |
Sw8wvfZ6XQ | Learning to Decide with AI Assistance under Human-Alignment | https://openreview.net/forum?id=Sw8wvfZ6XQ | [
"Nina Corvelo Benz",
"Eleni Straitouri",
"Manuel Rodriguez"
] | Poster | null | It is widely agreed that when AI models assist decision-makers in high-stakes domains, they should communicate the confidence of their predictions. However, empirical evidence suggests that decision-makers often struggle to determine when to trust a prediction based solely on this communicated confidence. In this conte... | [] | null | 18,503 | 2605.12646 | title_snapshot |
G3WiNUM5h1 | Achieve Latency-Efficient Temporal-Coding Spiking LLMs via Discretization-Aware Conversion | https://openreview.net/forum?id=G3WiNUM5h1 | [
"Jinjie Fang",
"Tianxing Man",
"Xiao Du",
"Chengxun Jin",
"Haozhen Zhang",
"Yi Chang",
"Bin Gu"
] | Poster | null | Large language models (LLMs) have achieved remarkable success while introducing critical energy bottlenecks that challenge sustainable deployment. Spiking neural networks (SNNs) provide a promising approach for energy-efficient spiking LLMs via ANN-to-SNN (A2S) conversion. Among various spike coding methods, time-to-fi... | [] | null | 15,876 | null | null |
0ZZ3tyBGyc | Rethinking Gradient Approximation in Quantization: A Zeroth-Order Expectation Perspective | https://openreview.net/forum?id=0ZZ3tyBGyc | [
"Wenkang Wang",
"Jia Zhang",
"Zelin Wei",
"Dongxu Liu",
"Jiabao Li",
"Wanli Shi",
"Bin Gu"
] | Poster | null | As large language models (LLMs) continue to scale, quantization has become a key technique for efficient deployment. However, multi-bit quantization employs non-differentiable rounding, hindering gradient optimization. Existing methods rely on heuristic surrogate gradients (e.g., STE), which work empirically but lack a... | [] | null | 13,985 | null | null |
jUAayjmSnv | SPRING: Solver-guided Process Rewards for Novel Logical Reasoning Steps Generation | https://openreview.net/forum?id=jUAayjmSnv | [
"Muhammad Asif Ali",
"Mohammad Raza",
"Wenqing Wang",
"Huan wang"
] | Poster | null | Logical reasoning remains a major challenge for large language models (LLMs), particularly on structured problems that require precise constraint tracking, consistency preservation, and multi-step deduction. This challenge is especially acute for small-scale LLMs, which are more prone to producing inconsistent, redunda... | [] | null | 18,404 | null | null |
Zz2YaAr1JB | Beyond What Seems Necessary: Hidden Gains from Scaling Training-Time Reasoning Length under Outcome Supervision | https://openreview.net/forum?id=Zz2YaAr1JB | [
"Yihao Xue",
"Allan Zhang",
"Jianhao Huang",
"Amit Sahai",
"Baharan Mirzasoleiman"
] | Poster | null | Training LLMs to think and reason for longer has become a key ingredient in building state-of-the-art models that can solve complex problems previously out of reach. Recent efforts pursue this in different ways, such as RL fine-tuning to elicit long CoT or scaling latent reasoning through architectural recurrence. This... | [] | null | 28,521 | 2602.00927 | title_snapshot |
nVZ10zYT13 | Modeling Whole-Slide Images as Dynamic Tumor Microenvironment Fields | https://openreview.net/forum?id=nVZ10zYT13 | [
"Lei Wu",
"Jiashuai Liu",
"Di Zhang",
"Zhangpeng Gong",
"Yingkang Zhan",
"Yi Niu",
"Jiusong Ge",
"Chunze Yang",
"Kai Yi",
"Mireia Crispin-Ortuzar",
"Chen Li",
"Zeyu Gao"
] | Poster | null | Due to the gigapixel-scale nature of whole-slide images (WSIs), weakly supervised WSI analysis is commonly formulated as a multiple instance learning (MIL) problem, where patch-level features are aggregated into slide-level representations. However, diagnostic and prognostic evidence often arises from spatially coheren... | [] | null | 3,821 | 2609.34451 | title_snapshot |
5GGreyp28W | Optimizing Computational-Statistical Runtime for Wasserstein Distance Estimation | https://openreview.net/forum?id=5GGreyp28W | [
"Peter M Jacobs",
"Jeff M Phillips"
] | Poster | null | Squared Wasserstein distance is a frequently used tool to compare probability distributions. This distance is typically computed between empirical measures of size $n$ from two underlying random samples. Unfortunately, even in lower dimensional Euclidean space problems $\left( d \in \{2,3\} \right)$, Wasserstein distan... | [] | null | 18,528 | 2605.20122 | title_snapshot |
YfsyoH7YFs | A New Perspective on Target-Conditioned Structural Dynamics for Link Prediction in Dynamic Graphs | https://openreview.net/forum?id=YfsyoH7YFs | [
"YUANYUAN XU",
"Yin Chen",
"Yingxuan Li",
"Wenjie Zhang",
"Xuemin Lin",
"Ying Zhang"
] | Poster | null | Dynamic link prediction requires determining whether the historical interactions of a source node provide reliable structural support for a specific target. Existing target-conditioned methods capture such support through structural heuristics such as repeated interactions or co-neighbor overlap, but typically instanti... | [] | null | 13,012 | null | null |
8td6EGze5A | Efficient Pre-Training with Token Superposition | https://openreview.net/forum?id=8td6EGze5A | [
"Bowen Peng",
"Théo Gigant",
"Jeffrey Quesnelle"
] | Poster | null | Pre-training of Large Language Models is often prohibitively expensive and inefficient at scale, requiring complex and invasive modifications in order to achieve high data throughput. In this work, we present Token-Superposition Training (TST), a simple drop-in method that significantly improves the data throughput per... | [] | null | 24,484 | 2605.06546 | title_snapshot |
61Yi6KS7Ek | FrameRouter: Frame Budget Routing for Long-Video Understanding on Video-MME-v2 and Beyond | https://openreview.net/forum?id=61Yi6KS7Ek | [
"Bingjun Luo",
"Jialin Guo",
"Siqi Li"
] | Poster | null | Long-video question answering benchmarks like Video-MME-v2 typically evaluate systems under a uniform frame budget for every question, despite stark variation in the evidence each query actually demands. We analyze this mismatch on Video-MME-v2 and find that different question subsets prefer different frozen frame-budg... | [] | null | 25,634 | null | null |
dnysHFYeGx | Algorithmic Impact Reveals the Hidden Structure of Alignment | https://openreview.net/forum?id=dnysHFYeGx | [
"Zachary Wojtowicz",
"Michelle Si",
"Finale Doshi-Velez",
"Ariel Procaccia"
] | Poster | null | When an algorithm makes a decision affecting multiple people, we implicitly have a social choice problem: how should differing opinions about the right course of action be reconciled into a single outcome? We show that if we make welfare consequences of alignment a first-order consideration, this problem can be reformu... | [] | null | 18,327 | null | null |
EbyKAh8h40 | Diffusion Meta-Prompting and Steering for Generalizable Foundation Model Adaptation | https://openreview.net/forum?id=EbyKAh8h40 | [
"Deepak Sridhar",
"Yi Li",
"Kartikeya Bhardwaj",
"Shuangjun Liu",
"TAOTAO JING",
"Yuan Li",
"Shuai Zhang",
"Jiancheng Lyu",
"Dashan Gao",
"Nuno Vasconcelos"
] | Poster | null | Prompt learning is a popular parameter-efficient method for adapting foundation models, but learned prompts are typically task-specific and fail to generalize to new classes, domains, or compositions of tasks. In this paper, we introduce a $\textbf{Diffusion Meta-Prompt (DMP)}$ model , a framework that models the distr... | [] | null | 8,509 | 2610.11067 | title_snapshot |
aZomMEAGSq | DRIVE: Fine-tuning via Data Contribution- and Diversity-aware Weighting with Prior Regularization | https://openreview.net/forum?id=aZomMEAGSq | [
"qing liu",
"Xinrui Chen",
"Weiyao Zhu",
"Yi Du",
"Ou Wu"
] | Poster | null | Single-stage fine-tuning is pervasive, yet catastrophic forgetting persists when the source pre-training data is inaccessible. Data weighting affords a natural lever to modulate sample influence. Yet existing weighting methods lack a principled mechanism to explicitly arbitrate knowledge retention against new-task adap... | [] | null | 158 | null | null |
ctcKsexBkZ | Outcome-Based RL Provably Leads Transformers to Reason, but Only With the Right Data | https://openreview.net/forum?id=ctcKsexBkZ | [
"Yuval Ran-Milo",
"Yotam Alexander",
"Shahar Mendel",
"Nadav Cohen"
] | Poster | null | Transformers trained via Reinforcement Learning (RL) with outcome-based supervision can spontaneously develop the ability to generate intermediate reasoning steps (Chain-of-Thought). Yet the mechanism by which sparse rewards drive policy gradient to discover such systematic reasoning remains poorly understood. We addre... | [] | null | 23,361 | 2601.15158 | title_snapshot |
4D7dKjdRtk | A Memory Efficient Unified Algorithm for Online Learning of Linear Dynamical Systems | https://openreview.net/forum?id=4D7dKjdRtk | [
"Yuval Ran-Milo",
"Angelos Assos",
"Elad Hazan"
] | Poster | null | Motivated by the challenge of stabilizing unknown linear dynamical systems (LDS) from observations, we study the fundamental prerequisite of online prediction. Our goal is to achieve sublinear regret with a memory footprint that adapts to the intrinsic complexity of the dynamics rather than the full hidden-state dimens... | [] | null | 23,312 | 2607.02050 | title_snapshot |
A7ffZI8Yma | Learning What Evaluators Value: A Reliable Approach to Modeling Evaluator Preferences | https://openreview.net/forum?id=A7ffZI8Yma | [
"Madeline Kitch",
"Nihar Shah"
] | Poster | null | In many machine learning applications, human and LLM evaluators use assessments of relevant criteria to create an overall evaluation for an item or individual. In applications like admissions, committees assess candidates on attributes such as test scores, GPA, and research experience to evaluate their overall fit for ... | [] | null | 17,463 | 2605.16615 | title_snapshot |
iQ0TBYaLes | Temporal-Scale Sensitivity in Time-Series Tokenization and Scale-Robust Token Estimation by Gated Sum | https://openreview.net/forum?id=iQ0TBYaLes | [
"Berken Utku Demirel",
"Christian Holz"
] | Poster | null | Transformers for temporal sequences require mapping sampled signals into sequences of vectors, called tokens. This is done by partitioning the sequence into local temporal windows (patches). We show that the tokenization step induces an approximation--estimation tradeoff. Smaller windows yield higher-variance statistic... | [] | null | 5,619 | null | null |
3fy5G5iNJf | On the Nonlinearity of Learning Rate Scaling for LLM Training | https://openreview.net/forum?id=3fy5G5iNJf | [
"ZAIWEN YANG",
"Huaqing Zhang",
"Jing Xu",
"Jingzhao Zhang"
] | Spotlight | null | Learning-rate transfer is critical for reducing the cost of training large language models: instead of sweeping learning rates at target scale, practitioners extrapolate from smaller runs. Existing approaches often assume that the optimal learning rate follows a log-linear scaling law in data scale and model size. We c... | [] | null | 31,053 | 2606.29158 | title_snapshot |
iQpv99SOhF | Self-Improvement Imitation with Biologically Guided Search for Protein Design Under Oracle Budgets | https://openreview.net/forum?id=iQpv99SOhF | [
"Ashima Khanna-Reiter",
"Dominik G Grimm"
] | Poster | null | Protein sequence optimization under tight oracle budgets requires methods that explore vast combinatorial spaces while making each evaluation informative. Existing reinforcement learning and off-policy generative approaches often degrade under surrogate noise, and position-agnostic mutation proposals risk disrupting fu... | [] | null | 18,015 | 2605.26690 | title_snapshot |
Vuh3ZNdgF6 | Trajectory-Consistent Dropout for Uncertainty Decomposition in Hamiltonian Neural Networks | https://openreview.net/forum?id=Vuh3ZNdgF6 | [
"Stephen J Roberts",
"Yuki Tachibana"
] | Poster | null | Hamiltonian Neural Networks (HNNs) produce stable long-horizon rollouts by integrating a learned Hamiltonian with a symplectic scheme. Standard Monte Carlo (MC) dropout conflicts with this per-trajectory interpretation because resampling the dropout mask inside a leapfrog integrator means successive substeps follow dif... | [] | null | 26,354 | null | null |
oqLzCT1cus | A Geometric Perspective on Reward Function Updates in Inverse Reinforcement Learning | https://openreview.net/forum?id=oqLzCT1cus | [
"Anish Abhijit Diwan",
"Jan Peters",
"Oleg Arenz"
] | Poster | null | Inverse Reinforcement Learning (IRL) is classically solved using Lagrangian optimization, by jointly optimizing a primal-dual problem with gradient descent to obtain the optimal reward function parameters and corresponding optimal policy. While this algorithmic view of IRL has been highly influential, gradient-based re... | [] | null | 34,814 | 2610.14041 | title_snapshot |
26nRwIZFcb | TrajWiki: Source-Grounded Memory Trajectories for Long-Horizon Dialogue Agents | https://openreview.net/forum?id=26nRwIZFcb | [
"Jingyu Sun",
"Yuyang Xue",
"Mingyang Li",
"Zhengtao Yao",
"Jiachen Li",
"Yang Cui",
"Wenhao Cai",
"Haozhe Liu",
"Fangying Wang",
"Magdalene K Montgomery",
"Syed M Baker",
"Hongpeng Zhou"
] | Poster | null | Large language model agents have shown strong capabilities in generating coherent and contextually appropriate responses, yet robust long-horizon dialogue remains limited by the lack of external memory that is traceable, updatable, and diagnostically transparent. Existing memory-augmented agents often store memories as... | [] | null | 16,943 | 2608.00967 | title_snapshot |
w2X7ALXCZx | Metropolis-Adjusted Diffusion Models | https://openreview.net/forum?id=w2X7ALXCZx | [
"Kevin H. Lam",
"Tyler Farghly",
"Christopher Williams",
"Jun Yang",
"Yee Whye Teh",
"Arnaud Doucet"
] | Poster | null | Sampling from score-based diffusion models incurs bias due to both time discretisation and the approximation of the score function. A common strategy for reducing this bias is to apply corrector steps based on the unadjusted Langevin algorithm (ULA) at each noise level within a predictor-corrector framework. However, U... | [] | null | 25,965 | 2605.09654 | title_snapshot |
SsF2kYJfYC | Is Decentralized LLM Agent RL Robust to Heterogeneity? An Asymmetric Tale | https://openreview.net/forum?id=SsF2kYJfYC | [
"Canyu Chen",
"Kangyu Zhu",
"Zhaorun Chen",
"Zhanhui Zhou",
"Shizhe Diao",
"Yiping Lu",
"Tian Li",
"Manling Li",
"Dawn Song"
] | Poster | null | Training AI agents powered by Large Language Models (LLMs) typically requires centralized access to user data, raising privacy and scalability concerns. We explore FedAgent, a decentralized reinforcement learning paradigm that collaboratively trains LLM agents across distributed clients without sharing local data. The ... | [] | null | 2,308 | null | null |
baoEzZg5LH | CoRE-RL: Co-evolving Reasoning Trajectories and Evidence Subgraphs with Learned Evidence Projection | https://openreview.net/forum?id=baoEzZg5LH | [
"Zhengquan luo",
"Guy Tadmor",
"Peilin Zhao",
"David Zeevi",
"Zhiqiang Xu"
] | Poster | null | Large language models have made knowledge-graph question answering more flexible, yet complex multi-hop questions often require evidence that becomes identifiable only during reasoning. Starting from the question alone, narrow retrieval can miss bridge facts, while broad expansion injects distractors and worsens long-c... | [] | null | 1,190 | null | null |
1BZdDrOyJH | SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning | https://openreview.net/forum?id=1BZdDrOyJH | [
"Peng Xia",
"Jianwen Chen",
"Hanyang Wang",
"Jiaqi Liu",
"Kaide Zeng",
"Yu Wang",
"Siwei Han",
"Yiyang Zhou",
"Xujiang Zhao",
"Haifeng Chen",
"Zeyu Zheng",
"Cihang Xie",
"Huaxiu Yao"
] | Poster | null | Large Language Model (LLM) agents have shown stunning results in complex tasks, yet they often operate in isolation, failing to learn from past experiences. Existing memory-based methods primarily store raw trajectories, which are often redundant and noise-heavy. This prevents agents from extracting high-level, reusabl... | [] | null | 9,850 | 2602.08234 | title_snapshot |
agLR3seGuO | GazeFlow: From Human Gaze Behavior to Generative Egocentric Gaze Prediction | https://openreview.net/forum?id=agLR3seGuO | [
"Sheng Zhao",
"Weikai Lin",
"Yuhao Zhu"
] | Poster | null | Egocentric gaze prediction enables many downstream applications but remains challenging, as human gaze is inherently stochastic. This stochasticity is constrained by structured temporal dynamics alternating between fixations and saccades, top-down influences from tasks, and bottom-up visual saliency. Based on this obse... | [] | null | 4,152 | 2609.38519 | title_snapshot |
IyJWTZzLb3 | Multidimensional Observer Model and Perceptual Dimensions of Human Image Quality Assessment | https://openreview.net/forum?id=IyJWTZzLb3 | [
"Sheng Zhao",
"Weikai Lin",
"Yuhao Zhu"
] | Poster | null | Judging image quality is not only ecologically relevant to everyday humans tasks, but also underpins many machine vision tasks such as image generation. This paper proposes a framework to understand the inherent perceptual space underlying image quality judgment in humans. We propose a multi-dimensional observer model ... | [] | null | 4,153 | 2609.38487 | title_snapshot |
KJ3CT9VwVV | L2P: Unlocking Latent Potential for Pixel Generation | https://openreview.net/forum?id=KJ3CT9VwVV | [
"Zhennan Chen",
"Junwei Zhu",
"Xu Chen",
"Jiangning Zhang",
"Jiawei Chen",
"Zhuoqi Zeng",
"Wei Zhang",
"Chengjie Wang",
"Jian Yang",
"Ying Tai"
] | Spotlight | null | Pixel diffusion models have recently regained attention for visual generation. However, training advanced pixel-space models from scratch demands prohibitive computational and data resources. To address this, we propose the \textbf{Latent-to-Pixel (L2P)} transfer paradigm, an efficient framework that directly harnesses... | [] | null | 4,346 | 2605.12013 | title_snapshot |
TjhQ07qEGv | A Constrained Bi-level Optimization Framework for Constrained Preference-Based Reinforcement Learning | https://openreview.net/forum?id=TjhQ07qEGv | [
"Yue Mao",
"Siyuan Xu",
"Shicheng Liu",
"Minghui Zhu"
] | Poster | null | This paper studies the problem of jointly learning a reward function, a cost function, and a policy from an expert's preferences. We formulate the problem as a constrained bi-level optimization problem, where the upper level infers the reward and cost functions from preferences, while the lower level optimizes a policy... | [] | null | 15,459 | null | null |
8oUdUwcYqg | STEP: Learning STructured Embeddings for Progressive Time Series | https://openreview.net/forum?id=8oUdUwcYqg | [
"Lucas Thil",
"Jesse Read",
"Rim Kaddah",
"Guillaume Doquet"
] | Poster | null | We present a novel method for learning interpretable representations of progressive time series, that is, data capturing irreversible state transitions such as degradation or task completion. Our approach uses a self-supervised contrastive objective to learn a low-dimensional latent space where progression manifests al... | [] | null | 25,378 | 2605.31061 | title_snapshot |
0pYlkhyLTf | Model-based Bootstrap of Controlled Markov Chains | https://openreview.net/forum?id=0pYlkhyLTf | [
"Ziwei Su",
"Imon Banerjee",
"Diego Klabjan"
] | Poster | null | We propose and analyze a model-based bootstrap for transition kernels in finite controlled Markov chains (CMCs) with possibly nonstationary or history-dependent control policies, a setting that arises naturally in offline reinforcement learning (RL) when the behavior policy generating the data is unknown. We establish ... | [] | null | 12,444 | 2605.12410 | title_snapshot |
oxoJ8MqY3p | Trie-Aware Transformers for Generative Recommendation | https://openreview.net/forum?id=oxoJ8MqY3p | [
"Zhenxiang Xu",
"Sirui Chen",
"Yong He",
"Jieyu Yang",
"Chuan Yuan",
"Ke Ding",
"Jing Cao",
"Can Wang",
"Jiawei Chen"
] | Poster | null | Generative recommendation (GR) aligns with advances in generative AI by casting next-item prediction as token-level generation rather than score-based ranking. Most GR methods adopt a two-stage pipeline: (i) \textit{item tokenization}, which maps each item to a sequence of discrete, hierarchically organized tokens; and... | [] | null | 24,317 | 2602.21677 | title_snapshot |
cumW3H2I4l | Commit Then Explore: Reasoning Models Diversify, Not Converge | https://openreview.net/forum?id=cumW3H2I4l | [
"Ben Jenkins",
"Mihaela Cardei"
] | Poster | null | Test-time compute scaling, where language models "think longer" via extended chain-of-thought (CoT), is the dominant paradigm for improving reasoning. Prevailing intuition holds that extended reasoning *converges* toward an answer; we show the opposite. Across three reasoning-RL families (DeepSeek-R1, Qwen-QwQ, Llama-3... | [] | null | 18,029 | null | null |
TToxD2UYY6 | Lattice Deduction Transformers | https://openreview.net/forum?id=TToxD2UYY6 | [
"Liam Davis",
"Alberto Alfarano",
"Leopold Haller",
"Mark Santolucito"
] | Poster | null | We introduce the Lattice Deduction Transformer (LDT), a recurrent transformer that approximates logically sound deduction by projecting its latent state through a lattice between forward passes. We train on-policy in a process that mirrors deduction in a search-based constraint solver and supervise training via a domai... | [] | null | 890 | 2605.08605 | title_snapshot |
34NGCK5xuh | Actor-Accelerated Policy Dual Averaging for Reinforcement Learning in Continuous Action Spaces | https://openreview.net/forum?id=34NGCK5xuh | [
"Ji Gao",
"Caleb Ju",
"Guanghui Lan",
"Zhaohui Tong"
] | Poster | null | Policy Dual Averaging (PDA) offers a principled Policy Mirror Descent (PMD) framework that more naturally admits value function approximation than standard PMD, enabling the use of approximate advantage (or Q-) functions while retaining strong convergence guarantees. However, applying PDA in continuous state and action... | [] | null | 14,859 | 2603.10199 | title_snapshot |
rqxWoOlmvG | ReCoVer: Resilient LLM Pre-Training System via Fault-Tolerant Collective and Versatile Workload | https://openreview.net/forum?id=rqxWoOlmvG | [
"Ziyue (Alvin) Liu",
"Zhengyang Wang",
"Ruijie (Ray) Zhang",
"Avinash Maurya",
"Hui Zhou",
"Paul Hovland",
"Sheng Di",
"Franck Cappello",
"Bogdan Nicolae",
"Zheng Zhang"
] | Poster | null | Pre-training large language models on massive GPU clusters has made hardware faults routine rather than rare, driving the need for resilient training systems. Yet existing frameworks either focus on specific parallelism schemes or risk drifting away from a failure-free training trajectory. We propose ReCoVer, a resilie... | [] | null | 7,315 | 2605.11215 | title_snapshot |
0QJyU7Bb8Y | Reading Attribution from Attention: Evidence Heads as Latent Attribution Mechanisms in LLMs | https://openreview.net/forum?id=0QJyU7Bb8Y | [
"Qi Wu",
"Jianfeng Qu",
"Peng-Fei Zhang",
"Yanzhe Ji",
"Siyu Li",
"Wei Chen",
"Zhixu Li",
"Jiajie Xu"
] | Poster | null | Large language models (LLMs) show strong performance in multi-document question answering, but their practical deployment is limited by unreliable and unfaithful attribution to supporting evidence. Existing prompting and training-based methods often suffer from hallucinated citations and lack interpretability in how ev... | [] | null | 16,420 | null | null |
avlwpkoCiV | World Models as Adversaries: Multi-Agent Self-Play Fine-Tuning for Robust Motion Planning | https://openreview.net/forum?id=avlwpkoCiV | [
"Tong Nie",
"Yuewen Mei",
"Junlin He",
"Yihong Tang",
"Jian Sun",
"Wei Ma"
] | Poster | null | Robust motion planning in dense traffic requires autonomous vehicles to interact in rare and safety-critical scenarios that are underrepresented in naturalistic driving data. Although adversarial training offers a feasible solution, existing methods often rely on external scenario generators, heuristic perturbations, o... | [] | null | 7,828 | 2607.10630 | title_snapshot |
zWigIoVsHo | Ensembling Language Models with Sequential Monte Carlo | https://openreview.net/forum?id=zWigIoVsHo | [
"Robin Chan",
"Tianyu Liu",
"Samuel Kiegeland",
"Clemente Pasti",
"Jacob Vigly",
"Timothy O'Donnell",
"Ryan Cotterell",
"Tim Vieira"
] | Poster | null | Combining the predictions of multiple models is one of the most effective strategies in machine learning. Thus, it is only natural to investigate the ensembling of language models (LMs). However, because LMs define distributions over strings, a proper ensemble requires aggregating predictions at the string level, which... | [] | null | 29,314 | 2603.05432 | title_snapshot |
548NwQAdhQ | Riemannian Lyapunov Framework: Optimization as Closed-Loop Control on Riemannian Manifolds | https://openreview.net/forum?id=548NwQAdhQ | [
"Yixuan Wang",
"Omkar Sudhir Patil",
"Warren Dixon"
] | Poster | null | We introduce a unified framework that reinterprets first order optimization on a smooth parameter space as a closed loop controlled dynamical system on a Riemannian manifold. Within this framework, optimizers are specified by a quadruple $(M, g, \Phi, \eta)$ consisting of a Riemannian metric, an internal state transiti... | [] | null | 22,263 | null | null |
Gteb9OmhzF | Corrective Diffusion Language Models | https://openreview.net/forum?id=Gteb9OmhzF | [
"Shuibai Zhang",
"Fred Peng",
"Yiheng Zhang",
"Jin Pan",
"Grigorios Chrysos"
] | Poster | null | While Diffusion Language Models (DLMs) are theoretically well-suited for iterative refinement due to their non-causal structure, off-the-shelf masked diffusion language models (MDLMs) often fail to reliably revise incorrect tokens in practice. The key challenge lies in the model's inability to distinguish between corre... | [] | null | 26,971 | 2512.15596 | title_snapshot |
cyktRcY8D7 | Learning from Trials and Errors: Reflective Test-Time Planning for Embodied LLMs | https://openreview.net/forum?id=cyktRcY8D7 | [
"Yining Hong",
"Huang Huang",
"Manling Li",
"Fei-Fei Li",
"Leonidas Guibas",
"Jiajun Wu",
"Yejin Choi"
] | Poster | null | Embodied LLMs endow robots with high-level task reasoning, but they cannot reflect on what went wrong or why, turning deployment into a sequence of independent trials where mistakes repeat rather than accumulate into experience. Drawing upon human reflective practitioners, we introduce Reflective Test-Time Planning, wh... | [] | null | 5,798 | 2602.21198 | title_snapshot |
VSnGy8znmQ | Not all uncertainty is alike: volatility, stochasticity, and exploration | https://openreview.net/forum?id=VSnGy8znmQ | [
"Payam Piray"
] | Poster | null | Adaptive decision-making in biological and artificial intelligence requires balancing the exploitation of known outcomes with the exploration of uncertain alternatives. Although prior work suggests that uncertainty generally promotes exploration, it has typically treated distinct sources of environmental uncertainty as... | [] | null | 13,180 | 2605.19215 | title_snapshot |
Vi8z1QsK4V | Vision Correlators: Correlation-Driven Visual Understanding with Hypergraphs | https://openreview.net/forum?id=Vi8z1QsK4V | [
"Mengqi Lei",
"Guohuan Xie",
"Siqi Li",
"Shihui Ying",
"Shaoyi Du",
"Jun-Hai Yong",
"Nanning Zheng",
"Yue Gao"
] | Poster | null | Modern visual backbones, despite their architectural diversity, largely reduce relational reasoning to pairwise message passing on graphs and therefore primarily capture 2-order interactions. However, visual understanding often depends on latent yet crucial higher-order semantic correlations among multiple regions, whi... | [] | null | 428 | null | null |
q45V4tagJs | Hydra: Towards Transferable Multi-Task Learning on Temporal Graphs | https://openreview.net/forum?id=q45V4tagJs | [
"Kiarash Shamsi",
"Farimah Poursafaei",
"Tran Gia Bao Ngo",
"Reihaneh Rabbany",
"Baris Coskunuzer",
"Guillaume Rabusseau",
"Michael Bronstein",
"Shenyang Huang",
"Cuneyt Akcora"
] | Poster | null | Real-world evolving networks are naturally modeled as temporal graphs (TGs), where capturing temporal dynamics is essential for predicting future graph properties that support downstream decision-making. Existing temporal graph methods have been developed primarily for single-task prediction, and little is known about ... | [] | null | 6,395 | null | null |
myI2BmrxeL | Fairness Failure Modes of Multimodal LLMs | https://openreview.net/forum?id=myI2BmrxeL | [
"Canyu Chen",
"Anglin Cai",
"Joan Nwatu",
"Yale Li",
"Jessica Hullman",
"Rada Mihalcea",
"Kathleen McKeown",
"Manling Li"
] | Poster | null | Although Multimodal Large Language Models (MLLMs) are increasingly deployed in high-stakes domains, the fairness of their outputs is under-explored. Building on the BBQ language bias benchmark, we construct a new dataset MultiBBQ using attested social biases and AI-generated photorealistic images for controllable fairn... | [] | null | 901 | null | null |
xMWhDplFaU | Rennala-NSGD: Asynchronous Stochastic Optimization Beyond Euclidean Geometry | https://openreview.net/forum?id=xMWhDplFaU | [
"Igor Sokolov",
"Alexander Tyurin",
"Peter Richtarik"
] | Poster | null | The recent empirical success of non-Euclidean optimizers, such as Muon (Jordan et al., 2024) and Shampoo (Gupta et al., 2018), has transformed the training of Large Language Models (LLMs) by exploiting the geometry of matrix spaces. Despite this progress, the theoretical understanding of stochastic optimization in gene... | [] | null | 27,317 | null | null |
L1cO6MnwfR | Scratchpad Patching: Decoupling Compute from Patch Size in Byte-Level Language Models | https://openreview.net/forum?id=L1cO6MnwfR | [
"Lin Zheng",
"Vasilisa Bashlovkina",
"Timothy Dozat",
"Dan Garrette",
"Laura Rimell",
"Joshua Maynez"
] | Poster | null | Tokenizer-free language models eliminate the tokenizer step of the language modeling pipeline by operating directly on bytes; patch-based variants further aggregate contiguous byte spans into patches for efficiency. However, the average patch size chosen at the model design stage governs a tight trade-off: larger patch... | [] | null | 30,616 | 2605.09630 | title_snapshot |
1XyUmZsyBA | Conservative neural posterior estimation via distributionally robust training | https://openreview.net/forum?id=1XyUmZsyBA | [
"William Laplante",
"Yuga Hikida",
"Charita Dellaporta",
"Francois-Xavier Briol",
"Ayush Bharti"
] | Poster | null | Simulation-based inference with neural posterior estimation (NPE) often yields overconfident and unreliable posteriors under limited simulation budgets. To address this, we propose DRO-NPE, a distributionally robust approach that replaces the standard NPE objective with a worst-case loss over a Wasserstein ambiguity se... | [] | null | 198 | 2605.28516 | title_snapshot |
Aak239DsQk | Horizon Adaptive Offline Policy Learning via Value Stitching | https://openreview.net/forum?id=Aak239DsQk | [
"Kexin ZHENG",
"Xianyuan Zhan",
"Xintao Yan"
] | Poster | null | Learning accurate value functions plays a decisive role for reinforcement learning (RL) agents to solve long-horizon, complex tasks. Conventional temporal-difference (TD) learning objectives suffer from value-estimation bias that accumulates over the horizon, while extended-horizon modeling methods, such as $n$-step TD... | [] | null | 8,543 | 2606.21136 | title_snapshot |
G7fOG13nTJ | Frequency‑Aware Flow Matching for Continuous and Consistent Robotic Action Generation | https://openreview.net/forum?id=G7fOG13nTJ | [
"Jianing Guo",
"Fangzheng Chen",
"Zihao Mao",
"WONG L Kenny",
"Zhenhong Wu",
"Yu Li",
"Yishuai Cai",
"Yuanpei Chen",
"Yikun Ban",
"Kai Chen",
"DOU QI",
"Yaodong Yang",
"Xianglong Liu",
"Huijie Zhao",
"Simin Li"
] | Poster | null | Flow matching has emerged as a standard paradigm for robotic manipulation owing to its strong expressive power for modelling complex, multimodal action distributions, alongside similar approaches like diffusion policy. However, existing methods rely on discretized action chunks, making them brittle to demonstrations co... | [] | null | 8,382 | 2606.20135 | title_snapshot |
yWb30l4iXg | Loyalty Capture: Reporting Relationships and Structural Sycophancy in Frontier AI Models | https://openreview.net/forum?id=yWb30l4iXg | [
"Eric So",
"Alex Imas"
] | Poster | null | We show that frontier AI models exhibit loyalty capture---a form of structural sycophancy in which a reporting relationship to a manager with misaligned incentives is sufficient to bias the model's recommendations, without any explicit instruction, user feedback, or adversarial prompting. In 5,757 trials across eight m... | [] | null | 12,498 | null | null |
geyjtAQ0PC | Policy Optimization in Tabular MDPs: Data-dependent Regret under Unknown Transitions | https://openreview.net/forum?id=geyjtAQ0PC | [
"Mingyi Li",
"Taira Tsuchiya",
"Kenji Yamanishi"
] | Poster | null | We study policy optimization for online episodic tabular Markov decision processes with unknown transition kernels, aiming for best-of-both-worlds guarantees and multiple data-dependent regret bounds. Recent work (Dann et al., 2023; Li et al., 2026) has shown that policy optimization can adapt to both adversarial and s... | [] | null | 11,057 | null | null |
9ifoNWFpH3 | Iterative Latent Refinement for Value Learning in Offline Goal-Conditioned RL | https://openreview.net/forum?id=9ifoNWFpH3 | [
"Daoxin Li",
"Songcheng Xu",
"Guozhang Chen"
] | Poster | null | Offline goal-conditioned reinforcement learning (GCRL) enables goal-reaching from static datasets, but the learned value, critic, or compatibility score must support both long-horizon reachability inference and offline policy extraction. We study iterative latent refinement as an architecture-level replacement for the ... | [] | null | 13,890 | null | null |
DSKSg1Kz1q | VSPO: Vector-Steered Policy Optimization for Controllable Model Behavior | https://openreview.net/forum?id=DSKSg1Kz1q | [
"Xuechen Zhang",
"Zijian Huang",
"Kai Yang",
"Weijia Zhang",
"Jiasi Chen",
"Samet Oymak"
] | Poster | null | Modern language models often need to optimize a primary accuracy objective while also accommodating secondary behavioral preferences, such as verbosity, agreeableness, or the level of technical expertise in its response. In practice, a base model may exhibit a desired behavior very rarely or not at all. Thus, endowing ... | [] | null | 10,649 | null | null |
TWF54YaFwi | Modelling Opinion Dynamics at Scale with Deep MARL | https://openreview.net/forum?id=TWF54YaFwi | [
"Lukas Seier",
"Brandon Kaplowitz",
"Sebastian Towers",
"Richard M Bailey",
"Jakob Foerster"
] | Poster | null | Modelling opinion dynamics typically relies on hand-crafted local interaction rules to study emergent macroscopic phenomena such as consensus and polarisation. In contrast, multi-agent reinforcement learning (MARL) enables agents to learn such behaviours directly by optimising simple rewards. To explore the potential o... | [] | null | 29,413 | 2606.07487 | title_snapshot |
BfBnHowvB9 | PhyMo: Learning Physical Dynamics with Accurate and Continuous Motion from Multi-View Videos | https://openreview.net/forum?id=BfBnHowvB9 | [
"Shangjia Liu",
"Jinxi Li",
"Siyuan Zhou",
"Bo Yang",
"Bing WANG"
] | Poster | null | Recovering dynamic 3D scene geometry, appearance, and the physical motion states that govern scene evolution from multi-view videos is important yet challenging. A key difficulty is that appearance can be deceptive. When trained primarily with rendering supervision, existing methods may fit the observed appearance well... | [] | null | 13,964 | null | null |
x2ON3LiaLg | Heterogeneous Agent Collaborative Reinforcement Learning | https://openreview.net/forum?id=x2ON3LiaLg | [
"Zhixia Zhang",
"Zixuan Huang",
"Gonxun Li",
"Huaiyang Wang",
"Chengyi Yuan",
"Xin Xia",
"deqing wang",
"Fuzhen Zhuang",
"Shuai Ma",
"Ning Ding",
"Yaodong Yang",
"Yikun Ban"
] | Poster | null | We introduce Heterogeneous Agent Collaborative Reinforcement Learning (HACRL), a new Reinforcement Learning from Verifiable Reward (RLVR) problem that addresses the inefficiencies of isolated multi-agent on-policy optimization. HACRL enables collaborative optimization with independent execution: heterogeneous agents sh... | [] | null | 10,834 | 2603.02604 | title_snapshot |
mwnXqnBf8n | FoundSurface: Feedforward Ray-Based 3D Scene Surface Reconstruction from Unposed Images | https://openreview.net/forum?id=mwnXqnBf8n | [
"Shenxing Wei",
"Jinxi Li",
"Zihui Zhang",
"Ajay Kumar",
"Wei Wang",
"Bo Yang"
] | Poster | null | Accurate 3D surface reconstruction from unposed images remains a fundamental challenge in computer vision. Recent feed-forward pointmap methods eliminate the need for camera poses but inherently produce discrete, sparse representations. In this paper, we present a generalizable pipeline that directly recovers continuou... | [] | null | 9,433 | null | null |
eBccMC00vY | From Density Matrices to Phase Transitions in Deep Learning: Spectral Early Warnings and Interpretability | https://openreview.net/forum?id=eBccMC00vY | [
"Max Hennick",
"Guillaume Corlouer"
] | Poster | null | A key problem in the modern study of Deep Learning is predicting and understanding emergent capabilities in models during training. Inspired by methods for studying reactions in quantum chemistry, we present the ``2-datapoint reduced density matrix" (2RDM). We show that this object provides a basis for computationally ... | [] | null | 22,607 | 2603.29805 | title_snapshot |
apl2nYU8FS | Adaptive and Neutral Theory of Evolution Strategies for Large Language Models | https://openreview.net/forum?id=apl2nYU8FS | [
"Soma Yokoi",
"Issei Sato"
] | Poster | null | Evolution strategies (ES) have recently emerged as a promising alternative to gradient-based reinforcement learning for fine-tuning large language models, but classical zeroth-order optimization theory does not systematically account for the empirical observations driving this success. We formulate the continuous-time ... | [] | null | 25,331 | null | null |
MXIzIx8qk4 | Dialectics of Alignment: Harnessing Unsafe Knowledge for Dynamic Safety Routing | https://openreview.net/forum?id=MXIzIx8qk4 | [
"Maryam Hashemzadeh Barvarz",
"Jerry Huang",
"Minseon Kim",
"Marc-Alexandre Côté",
"Sarath Chandar"
] | Poster | null | The prevailing paradigm in large language model (LLM) alignment operates via erasure, filtering unsafe data or training models to strictly refuse harmful prompts. While effective at reducing immediate toxicity, this approach fundamentally constricts the model's epistemological scope, resulting in over-cautious systems ... | [] | null | 20,477 | 2606.00686 | title_snapshot |
cbLnr5GrNf | Complementing DINO Features with Image Structure for Part Discovery | https://openreview.net/forum?id=cbLnr5GrNf | [
"Samyak Rawlekar",
"Nikhil C Paleti",
"Amey Gupta",
"Narendra Ahuja"
] | Poster | null | Self-supervised Vision Transformers (ViTs) like DINO encode rich semantics, enabling training-free segmentation of objects from the background and discovery of their parts via feature clustering. This paper focuses on part discovery: partitioning a foreground object into semantic parts. Existing approaches produce impr... | [] | null | 14,979 | null | null |
kdC8R9ryni | MemPlan: Memory-Conditioned PDDL Planning for Partially Observable Text Environments | https://openreview.net/forum?id=kdC8R9ryni | [
"Tengwei Song",
"Robert Hoehndorf"
] | Poster | null | LLM agents are flexible in language interaction, but using them as long-horizon planners is costly and prone to hallucinated or invalid actions. This paper targets a different role for LLMs: constrained bridges between language and symbolic planning. We propose MemPlan, a memory-conditioned PDDL planning framework for ... | [] | null | 17,334 | null | null |
0vSuVjlB0R | Commutator Memory: Sparse, Path-Local Reading and Steering in Language Models | https://openreview.net/forum?id=0vSuVjlB0R | [
"John Sweeney"
] | Poster | null | Swapping two LLM post-training stages can change the final model, but an aggregate loss or benchmark delta only says that something moved, not where the order-dependent residue landed. We ask whether this residue leaves a memory trace: a structured signal that is localized in output space, changes held-out order-gap NL... | [] | null | 31,543 | 2609.34348 | title_snapshot |
Mvqk8cd5uy | Emergent Visual Thinking in Text-Only Reasoning through Multimodal Training | https://openreview.net/forum?id=Mvqk8cd5uy | [
"Qi Zhang",
"Bohan Wang",
"QIANRU SUN",
"Hanwang Zhang"
] | Poster | null | Does multimodal training make a language model think visually, even when the input is text only? We test this with targeted causal ablations across 10 multimodal configurations from 6 model families, with matched text-only backbones and cross-family transfer controls. We contrast spatial-imagery tasks that invite menta... | [] | null | 4,672 | null | null |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.