AI conference papers

AI conference paper library

Search papers from 30+ top conferences, including NeurIPS, ICLR, ICML, CVPR, ACL and AAAI. Filter by keyword, then open any paper for its abstract, authors and links.

ICML 2026

Process Reward Agents for Steering Knowledge-Intensive Reasoning

Jiwoong Sohn, Tomasz Sternal, Kenneth Styppa, Torsten Hoefler, Michael Moor

Reasoning in knowledge-intensive domains remains challenging as intermediate steps are often not locally verifiable: unlike math or code, evaluating step correctness may require synthesizing clues across large external knowledge sources. As a result, subtle errors can propagate through reasoning traces, potentially never to be detected. Prior work has proposed process reward models (PRMs), including retrieval-augmented variants, but these methods operate post hoc, scoring completed trajectories, which prevents their integration into dynamic inference procedures. Here, we introduce Process Reward Agents (PRA), an inference-time method for providing domain-grounded, online, step-wise rewards to a frozen policy. In contrast to prior retrieval-augmented PRMs, PRA enables search-based decoding to rank and prune candidate trajectories at every generation step. Experiments on multiple medical reasoning benchmarks demonstrate that PRA consistently outperforms strong baselines, achieving 81.9% accuracy on MedQA with Qwen3-4B, a new state of the art at the 4B scale. Importantly, PRA generalizes to unseen frozen policy models ranging from 0.5B to 8B parameters, improving their accuracy by up to 25.7% without any policy model updates. More broadly, PRA suggests a paradigm in which frozen reasoners are decoupled from domain-specific reward modules, allowing the deployment of new backbones in complex domains without retraining. All code and data are publicly available at https://process-reward-agents.github.io/.

ICML 2026

Bias-Spectrum Neural Processes for Parametric PDEs: Architecture Priors Meet PDE Constraints

Hui Li, Huafeng Liu, Chenguang Li, Tianxiao Zhang, Yajun Yang, Liping Jing

Parametric partial differential equations (PDEs) serve as fundamental models across science and engineering, yet constructing fast and accurate surrogate models from sparse, irregularly sampled observations with reliable uncertainty quantification remains challenging. Existing approaches struggle to simultaneously handle variable observation patterns, preserve physics consistency, and provide well-calibrated predictive uncertainty. We introduce Bias-Spectrum Neural Processes (BSNP), a unified meta-learning framework that systematically integrates weak structural priors (translation equivariance, locality) with strong physical priors (governing equations and boundary conditions). BSNP addresses two critical obstacles: discretization overfitting through stochastic collocation that resamples residual evaluation points, and uncertainty collapse through mean-field enforcement that applies PDE constraints only to predictive means while preserving learned uncertainty. Comprehensive experiments on nonlinear Poisson equations, Burgers dynamics, and Navier-Stokes flows demonstrate that BSNP achieves superior accuracy and well-calibrated uncertainty quantification in sparse-data regimes.

ICML 2026

Rethinking the Design Space of Reinforcement Learning for Diffusion Models: On the Importance of Likelihood Estimation Beyond Loss Design

Jaemoo Choi, Yuchen Zhu, Wei Guo, Petr Molodyk, Bo Yuan, Jinbin Bai, Yi Xin, Molei Tao, Yongxin Chen

Reinforcement learning has been widely applied to diffusion and flow models for visual tasks such as text-to-image generation. However, these tasks remain challenging because diffusion models have intractable likelihoods, which creates a barrier for directly applying popular policy-gradient type methods. Existing approaches primarily focus on crafting new objectives built on already heavily engineered LLM objectives, using ad hoc estimators for likelihood, without a thorough investigation into how such estimation affects overall algorithmic performance. In this work, we provide a systematic analysis of the RL design space by disentangling three factors: i) policy-gradient objectives, ii) likelihood estimators, and iii) rollout sampling schemes. We show that adopting an evidence lower bound (ELBO) based model likelihood estimator, computed only from the final generated sample, is the dominant factor enabling effective, efficient, and stable RL optimization, outweighing the impact of the specific policy-gradient loss functional. We validate our findings across multiple reward benchmarks using SD 3.5 Medium, and observe consistent trends across all tasks. Our method improves the GenEval score from $0.24$ to $0.95$ in $90$ GPU hours, which is $4.6\times$ more efficient than FlowGRPO and $2\times$ more efficient than the SOTA method DiffusionNFT without reward hacking.

ICML 2026

MechVQA: Benchmarking and Enhancing Multimodal LLMs on Comprehensive Mechanical Drawing Understanding

Qian Kou, Xiaofeng Shi, Yulin Li, Xiaosong Qiu, Xinyang Wang, Hua Zhou, Cao Dongxing

Multimodal Large Language Models (MLLMs) have demonstrated significant achievements in general visual question answering (VQA) tasks. However, they remain brittle on mechanical en- gineering drawings, where high annotation den- sity and weak domain knowledge, compounded by unreliable spatial relation reasoning under strict projection rules and geometric constraints, make decisive cues easy to miss and frequently lead to wrong answers. To bridge this gap, we introduce the first comprehensive mechani- cal drawing understanding dataset, MechVQA, created through a semi-automated construction and quality-control pipeline. MechVQA con- tains 3.3k high-density pictures with 21K ques- tion–answer pairs, spanning 10 different fine- grained tasks across three capability levels: Recognition, Reasoning, and Judging, provid- ing a testbed to evaluate and improve MLLM understanding on real-world mechanical draw- ings. On top of MechVQA, we then develop the MechVL model through a multi-stage training paradigm, building a strong domain-specialized baseline. Extensive experimental results demon- strate that MechVL outperforms the strongest closed-source baseline by 7.57 percentage points on the MechVQA total score, significantly en- hancing mechanical drawing understanding abil- ity and providing a reusable foundation for deploy- ing MLLMs in mechanical design and inspection scenarios.

ICML 2026

Distilling Task-Level Coordination Policies for Generalizable Multi-Agent Cooperation

Zimo Zhai, Manjie Xu, Wei Liang

Large language models have shown strong reasoning abilities and are increasingly explored as high-level coordinators for multi-agent systems. However, directly deploying LLMs for coordination remains challenging, as effective policies often fail to reliably emerge at the low-level control stage, and inference costs limit scalability. We propose SynCoord (Synthetic Coordination Distillation), a self-supervised pipeline that distills task-level decision-making for cooperation from high-capacity reasoning models into lightweight agent policies. Our approach does not rely on explicit supervision or handcrafted coordination rules. Instead, we define a set of task-level tool interfaces that constrain LLM interaction and enable the collection of interaction trajectories, which are then used to train compact coordinated policies. This distillation process transfers coordination behaviors that are difficult to elicit through prompting alone, while substantially reducing inference overhead at execution time. We evaluate our method on cooperative multi-agent benchmarks including Overcooked-AI and Level-Based Foraging (LBF), under varying team sizes and environment scales. Experimental results show that the distilled policies achieve success rates and execution efficiency comparable to reinforcement learning–based methods, while exhibiting fewer erroneous or redundant actions. Moreover, the learned task-level coordination policy generalizes effectively to unseen team compositions and larger layouts without retraining.

ICML 2026

Enhancing Cross-subject Emotion Recognition via Heterogeneous Distribution Augmentation and Collaborative Learning

Wending Xiong, Ruimin Hu, Lingfei Ren, Junhang Wu, Mei Wang, Dengshi Li, Mang Ye

Cross-subject emotion recognition aims to improve a model's generalization to previously unseen subjects. Existing methods are mainly built upon domain generalization or data augmentation, but suffer from two major limitations: 1) heavy dependence on modality-specific feature designs—almost exclusively tailored to EEG signals—resulting in limited generalizability; and 2) the widespread assumption of independently and identically distributed data, which restricts the diversity of generated samples. To address these challenges, we systematically analyze the heterogeneous distribution characteristics of emotion data and propose **MixEmo**, a framework that integrates heterogeneous distribution augmentation and collaborative learning. Specifically, a well-trained backbone is used to extract representations and partition them into multiple single-distribution subsets as distribution prototypes. These prototypes are randomly combined to synthesize unseen distributions, thereby enhancing distributional diversity. Finally, heterogeneous distribution collaborative learning jointly optimizes the model across subsets. Extensive experiments demonstrate that **MixEmo** substantially improves generalization performance in cross-subject emotion recognition.

ICML 2026

Accurate, Private, Secure, Federated U-statistics with Higher Degree

Quentin Sinh, Jan Ramon

We study the problem of computing a U-statistic with a kernel function $f$ of degree $k \geq 2$, i.e., the average of some function $f$ over all $k$-tuples of instances, in a federated learning setting. U-statistics of degree $2$ include several useful statistics such as Kendall's $\tau$ coefficient, the Area under the Receiver-Operator Curve and the Gini mean difference. Existing methods provide solutions only under the lower-utility local differential privacy model and/or scale poorly in the size of the domain discretization. In this work, we propose a protocol that securely computes U-statistics of degree $k \geq 2$ under central differential privacy by leveraging Multi Party Computation (MPC). Our method substantially improves accuracy when compared to prior solutions. We provide a detailed theoretical analysis of its accuracy, communication and computational properties. We evaluate its performance empirically, obtaining favorable results, e.g., for Kendall's $\tau$ coefficient, our approach reduces the Mean Squared Error by up to four orders of magnitude over existing baselines.

ICML 2026

Efficient and Minimax Optimal In-context Nonparametric Regression with Transformers

Michelle Ching, Ioana Popescu, Nico Smith, Tianyi Ma, William G. Underwood, Richard J. Samworth

We study in-context learning for nonparametric regression with $\alpha$-Hölder smooth regression functions, for some $\alpha>0$. We prove that, with $n$ in-context examples and $d$-dimensional regression covariates, a pretrained transformer with $\Theta(\log n)$ parameters and $\Omega(n^{2\alpha/(2\alpha+d)}\log^3 n)$ pretraining sequences can achieve the minimax optimal rate of convergence $O(n^{-2\alpha/(2\alpha+d)})$ in mean squared error. Our result requires substantially fewer transformer parameters and pretraining sequences than previous results in the literature. This is achieved by showing that transformers are able to approximate local polynomial estimators efficiently by implementing a kernel-weighted polynomial basis and then running gradient descent.

ICML 2026

Learning to Label: A Reinforced Self-Evolving Framework for Semi-supervised Referring Expression Segmentation

Runlong Cao, Ying Zang, Chuanwei Zhou, Tianrun Chen, Tong Zhang, Zhen Cui, Chunyan Xu

Semi-supervised referring expression segmentation (SS-RES) aims to achieve precise pixel-level language grounding under limited annotation, yet suffers from limited supervision and unreliable pseudo-labels when exploiting unlabeled image–text pairs. In this work, we propose Learning to Label, a reinforced self-evolving framework (L2L) that casts pseudo-label construction as a learnable decision-making process. To build foundational understanding, we leverage a multimodal large language model to extract semantic–spatial priors, which are instantiated as initial soft segmentation proposals and elevated—together with textual cues—into learnable guidance signals that condition a hierarchical segmentation network. To ensure stable learning, a reinforced pseudo-label selection is further formulated as an exploratory decision process that adaptively rewards high-utility pixel-level supervision based on multimodal priors and model predictions. This reinforced self-evolving loop enables joint optimization of the segmentation model and pseudo-labels, progressively enhancing label reliability under sparse supervision. Extensive experiments on RefCOCO, RefCOCO+, and RefCOCOg datasets demonstrate improvements over existing methods, validating its effectiveness and generalization.

ICML 2026

Mind-Omni: A Unified Multi-Task Framework for Brain-Vision-Language Modeling via Discrete Diffusion

Yizhuo Lu, Changde Du, Qingyu Shi, Hang Chen, Jie Peng, Liuyun Jiang, Shuangchen Zhao, Huiguang He

Modeling the interplay between external stimuli and internal neural representations is a pivotal research area for Brain-Computer Interfaces (BCIs). A major limitation of prior work is the prevailing paradigm of specialized, single-task models, which curtails versatility and neglects inter-task synergies. To address this, we propose Mind-Omni, the first versatile framework that unifies seven distinct encoding and decoding tasks through a discrete diffusion paradigm. At its core is a novel Brain Tokenizer that transforms heterogeneous, continuous brain signals into standardized, discrete tokens. This enables direct, token-level interactions for mutual understanding and generation between any two or more modalities within a shared semantic space. To unlock advanced reasoning capabilities, we further curate a specialized Brain Question Answering (BQA) instruction-tuning dataset. Our model not only establishes a new state-of-the-art among multi-task unified frameworks but also provides strong evidence for multi-task synergy. By demonstrating performance competitive with, and at times superior to, larger specialized models, our work offers a powerful new paradigm for neural modeling and paves the way for foundation models of neural activity.

ICML 2026

On Contraction of Sequential and Offset Rademacher Complexities

Adam Block, Alexander Rakhlin, Mark Sellke

The Rademacher complexity of a function class is among the most basic notions of its ``size'' and yields classical offline generalization bounds for Lipschitz loss functions that lead in turn to a modern understanding of statistical learning. More recently, the *sequential* and *offset* Rademacher complexities were introduced to prove analogous generalization bounds for online learning and for prediction with squared loss. A fundamental structural result in the theory of Rademacher complexity, with many applications to learning theory, is the Ledoux--Talagrand contraction lemma, which states that the Rademacher complexity of a composition of a function class with a fixed Lipschitz function is at most that of the original class. We show that, under structural assumptions on the function class, this contraction extends to sequential and offset Rademacher complexity at the price of polylogarithmic factors. We further show that these logarithmic factors cannot be removed in general and, absent these additional structural assumptions, no such contraction inequality can hold. These results together indicate that the sequential and offset Rademacher complexities behave fundamentally differently from the classical Rademacher complexity with respect to contraction, which in turn has broad implications for understanding the sample complexities of online learning and regression with squared loss for composed function classes.

ICML 2026

MAS-Orchestra: Understanding and Improving Multi-Agent Reasoning Through Holistic Orchestration and Controlled Benchmarks

Zixuan Ke, Yifei Ming, Austin Xu, Ryan Chin, Xuan-Phi Nguyen, Prathyusha Jwalapuram, Jiayu Wang, Semih Yavuz, Caiming Xiong, Shafiq Joty

While multi-agent systems (MAS) promise elevated intelligence through coordination of agents, current approaches to automatic MAS design under-deliver. Such shortcomings stem from two key factors: (1) methodological complexity – agent orchestration is performed using sequential, code-level execution that limits global system-level holistic reasoning and scales poorly with agent complexity – and (2) efficacy uncertainty – MAS are deployed without understanding if there are tangible benefits compared to single-agent systems (SAS). We propose MASOrchestra, a training-time framework that formulates MAS orchestration as a function-calling reinforcement learning problem with holistic orchestration, generating an entire MAS at once. In MAS-Orchestra, complex, goal-oriented subagents are abstracted as callable functions, enabling global reasoning over system structure while hiding internal execution details. To rigorously study when and why MAS are beneficial, we introduce MASBENCH, a controlled benchmark that characterizes tasks along five axes: Depth, Horizon, Breadth, Parallel, and Robustness. Our analysis reveals that MAS gains depend critically on task structure, verification protocols, and the capabilities of both orchestrator and subagents, rather than holding universally. Guided by these insights, MAS-Orchestra achieves consistent improvements on public benchmarks including mathematical reasoning, multi-hop QA, and search-based QA, while achieving more than 10× efficiency over strong baselines. Together, MAS-Orchestra and MASBENCH enable better training and understanding of MAS in the pursuit of multi-agent intelligence.

ICML 2026

Reasoning Models Struggle to Control their Chains of Thought

Chen Yueh-Han, Robert McCarthy, Bruce W. Lee, He He, Micah Carroll, Tomek Korbak

Instruction following in LLMs captures models' ability to change their visible behaviors as requested by users. Instead, we study models' ability to control their chain-of-thought (CoT). This capability -- CoT controllability -- is undesirable because it could allow models to suppress signs of misbehavior in their CoT, thereby undermining our ability to monitor them. To measure this, we introduce the \emph{CoT-Control} evaluation suite. We show that reasoning models are less able to follow instructions in their CoT than in their outputs: on instructions like reasoning about a genetics problem without mentioning the word ``chromosome", Claude-Sonnet-4.5 complies only 5\% of the time. We also find that CoT controllability is higher for larger models and decreases with more RL training, test-time compute, and increased problem difficulty. CoT controllability failures extend even to situations in which models are given incentives (as opposed to direct requests) to evade CoT monitors, although models that are told they're being monitored exhibit slightly higher controllability. Similarly, eliciting controllability by adversarially optimizing prompts doesn’t meaningfully increase controllability. Our results leave us cautiously optimistic: reasoning models generally seem characterized by low CoT controllability. However, the mechanism behind this phenomenon is not well understood. Given its importance for maintaining CoT monitorability, we recommend that frontier labs keep tracking controllability for future models.

ICML 2026

QiMeng-ChipV-RTL: Exploiting Information Locality for IP-level Verilog Generation

Hanqi Lyu, Di Huang, Yaoyu Zhu, Kangcheng Liu, Bohan Dou, Chongxiao Li, Pengwei Jin, Shuyao Cheng, Rui Zhang, Zidong Du, Qi Guo, Xing Hu, Yunji Chen

The generation of Register-Transfer Level (RTL) code is a crucial yet labor-intensive step in digital hardware design, traditionally requiring engineers to manually translate complex specifications into thousands of lines of synthesizable Hardware Description Language (HDL) code. While Large Language Models (LLMs) have shown promise in automating this process, existing approaches—including fine-tuned domain-specific models and advanced agent-based systems—struggle to scale to industrial IP-level design tasks. We identify three key challenges: (1) handling long, highly detailed documents, where critical interface constraints become buried in unrelated submodule descriptions; (2) generating long RTL code, where both syntactic and semantic correctness degrade sharply with increasing output length; and (3) navigating the complex debugging cycles required for functional verification through simulation and waveform analysis. To overcome these challenges, we propose \textit{ChipV-RTL}, a multi-agent framework that leverages \textit{information locality} in modular hardware design. ChipV-RTL decomposes the long-document to long-code generation problem into a set of short-document, short-code tasks, enabling scalable generation and debugging. Specifically, ChipV-RTL integrates hierarchical document partitioning, task planning, localized code generation, interface-consistent merging, and AST-guided locality-aware debugging. Experiments on \textsc{RealBench}, an IP-level Verilog generation benchmark, demonstrate that ChipV-RTL substantially outperforms state-of-the-art (SOTA) LLMs and agents, achieving a pass rate of 45.0\% compared to 21.6\%. Code, project page are available at: \url{https://iprc-dip.github.io/ChipV-RTL/}.

ICML 2026

Native Active Perception as Reasoning for Omni-Modal Understanding

Zhenghao Xing, Ruiyang Xu, Yuxuan Wang, Jinzheng He, Ziyang Ma, Qize Yang, Yunfei Chu, Jin Xu, Junyang Lin, Chi-Wing Fu, Pheng-Ann Heng

Passive models for long video understanding typically rely on a "watch-it-all" paradigm, processing frames uniformly regardless of query difficulty, causing computational cost to grow with video duration. Although interactive frameworks have emerged, they often rely on global pre-scanning, and their context cost still scales with video length. We propose **OmniAgent**, the first native omni-modal agent that formulates video understanding as a POMDP-based iterative **Observation-Thought-Action** cycle. OmniAgent executes on-demand actions to selectively distill audio-visual cues into a persistent textual memory, effectively decoupling reasoning complexity from raw video duration. To operationalize this, we introduce (1) **Agentic Supervised Fine-Tuning** to bootstrap native active perception via best-of-N trajectory synthesis with dual-stage quality control, and (2) **Agentic Reinforcement Learning** with **TAURA** (Turn-aware Adaptive Uncertainty Rescaled Advantage), which leverages turn-level entropy to steer credit assignment toward pivotal discovery turns. Crucially, OmniAgent exhibits positive test-time scaling, where performance improves as the number of reasoning turns increases, validating the efficacy of active perception. Empirical results across ten benchmarks (e.g., VideoMME, LVBench) demonstrate that OmniAgent achieves state-of-the-art performance among open-source models. Notably, on LVBench, our 7B agent outperforms the $10\times$ larger Qwen2.5-VL-72B (50.5% vs. 47.3%). We release our code and model at https://github.com/HarryHsing/OmniAgent.

ICML 2026

Capability Traps in DPO

Marco Pollanen

Direct Preference Optimization (DPO) is often tuned by selecting checkpoints with high preference margins. We show that this selection rule can fail. Across dense $\beta$ sweeps in three 7B open-weight families under controlled DPO recipes, the DPO preference margin can strongly anticorrelate with capability probes, with the strongest observed case in Llama logic probes (Pearson $r=-0.91$, $n=13$). In Mistral, matched-duration controls show that transient exposure to higher $\beta$ produces persistent arithmetic ($d_z = 3.73$, $p = 0.002$) and format ($d_z = 3.38$, $p = 0.003$) degradation relative to a duration-matched constant-$\beta$ run, while logic and sycophancy show no significant effect. An expanded 14-probe logic suite further shows that aggregate capability scores can hide stable opposition between probe clusters: direct-inference probes are consistently negative while fallacy-detection probes are consistently positive, with 13 of 14 probes sign-stable across 3 seeds and 3 $\beta$ values. Sensitivity profiles differ across architectures: Mistral shows elevated seed variance near $\beta \approx 10^{-2}$, Llama shows capability-specific rigidity, and Qwen trades off smoothly. These findings motivate probe-resolved $\beta$ sweeps rather than margin-based checkpoint selection.

ICML 2026

GLARE: Scalable Neuro-Symbolic Reward Shaping for LLM Agents via Group-Level Automata

Jingyuan Yan, Qingchen Liu, Qichao Ma, Jiahu Qin

Reinforcement Learning (RL) with Group Relative Policy Optimization (GRPO) shows great promise for enhancing LLM reasoning, but remains challenged by sparse and unstable rewards in long-horizon tasks. Existing approaches to reward shaping struggle to balance semantic expressiveness, reliability, and computational efficiency: heuristic rules lack flexibility, while LLM-as-a-Judge incurs high computational cost and suffer from inconsistent and misaligned scoring signals in long-context settings. To address these challenges, we introduce GLARE, a neuro-symbolic reward framework that decouples semantic abstraction from credit assignment. Specifically, to leverage semantic understanding while preserving symbolic determinism, we first extract and symbolize trajectory events into a discrete representation. These events are then translated into Linear Temporal Logic (LTL) formulas, which are compiled into deterministic automata that track the agent's progress via state transitions. This mechanism yields dense and consistent reward signals, avoiding unstable direct scoring while significantly reducing computational cost. Empirical results on ALFWorld show that GLARE outperforms GRPO by 12.1\% in success rate, while achieving an 8.1\% improvement over conventional LLM-based judges using only 15\% of their computational cost.

ICML 2026

HDFlow: Hierarchical Diffusion-Flow Planning for Long-horizon Tasks

Gireesh Nandiraju, Yuanliang Ju, Chaoyi Xu, Weiheng Liu, Yuxuan Wan, He Wang

Recent advances in generative models have shown promise in generating behavior plans for long-horizon, sparse reward tasks. While these approaches have achieved promising results, they often lack a principled framework for hierarchical decomposition and struggle with the computational demands of real-time execution, due to their iterative denoising process. In this work, we introduce $\textbf{Hierarchical Diffusion-Flow}$ ($\texttt{\textbf{HDFlow}}$), a novel hierarchical planning framework that optimally leverages the strengths of $\textit{diffusion}$ and $\textit{rectified flow}$ models to overcome the limitations of single-paradigm generative planners. $\texttt{\textbf{HDFlow}}$ employs a high-level diffusion planner to generate sequences of strategic subgoals in a learned latent space, capitalizing on diffusion's powerful exploratory capabilities. These subgoals then guide a low-level rectified flow planner that generates smooth and dense trajectories, exploiting the speed and efficiency of ordinary differential equation (ODE)-based trajectory generation. We evaluate $\texttt{\textbf{HDFlow}}$ on four challenging furniture assembly tasks in both simulation and real-world, where it significantly outperforms state-of-the-art methods. Furthermore, we also showcase our method's generalizability on two long-horizon benchmarks comprising diverse locomotion and manipulation tasks. Project website: https://hdflow-page.github.io/

ICML 2026

Extracting alignment data in open models

Federico Barbero, Xiangming Gu, Christopher A. Choquette-Choo, Chawin Sitawarin, Matthew Jagielski, Itay Yona, Petar Veličković, Ilia Shumailov, Jamie Hayes

In this work, we show that it is possible to extract significant amounts of alignment training data from a post-trained model -- useful to steer the model to improve certain capabilities such as long-context reasoning, safety, instruction following, and maths. While the majority of related work on memorisation has focused on measuring success of training data extraction through string matching, we argue that embedding models are better suited for our specific goals. Distances measured through a high quality embedding model can identify semantic similarities between strings that a different metric such as edit distance will struggle to capture. In fact, in our investigation, approximate string matching would have severely undercounted (by a conservative estimate of $10\times$) the amount of data that can be extracted due to trivial artifacts that deflate the metric. Interestingly, we find that models readily regurgitate training data that was used in post-training phases such as SFT or RL. We show that this data can be then used to train a base model, recovering a meaningful amount of the original performance. We believe our work exposes a possibly overlooked risk towards extracting alignment data. Finally, our work opens up an interesting discussion on the downstream effects of distillation practices: since models seem to be regurgitating aspects of their training set, distillation can therefore be thought of as indirectly training on the model's original dataset.

ICML 2026

Fix the Loss, Not the Radius: Rethinking the Adversarial Perturbation of Sharpness-Aware Minimization

Jinping Wang, Qinhan Liu, Zhiwu Xie, Zhiqiang Gao

Sharpness-Aware Minimization (SAM) improves generalization by minimizing the worst-case loss within a fixed parameter-space radius neighborhood. SAM and its variants mainly rely on a first-order linearized surrogate, while flat minima are inherently a second-order (curvature) notion. We revisit this mismatch and propose Loss-Equated SAM (LE-SAM), which inverts the traditional SAM mechanism that replaces the fixed perturbation radius with a fixed loss-space budget, effectively removing gradient-norm–dominated learning signals and shifting optimization toward curvature-dominated terms. Extensive experiments across diverse benchmarks and tasks demonstrate the strong generalization ability of LE-SAM that consistently outperforms SAM and even its variants, achieving state-of-the-art performance.

Keep browsing

More papers from top AI conferences

30+ top conferences, by conference, year, and topic.