ML code should run where researchers choose, not where it was first written. Today, machine learning code mainly lives in two ecosystems, PyTorch on NVIDIA GPUs and JAX on Google TPUs. Migrating repositories between them means building and testing across frameworks and hardware. This migration is (1) a practical need, since it unlocks different hardware for researchers, and (2) a natural benchmark: it involves different frameworks and hardware, covers whole repositories, and is long-horizon and challenging. We introduce XStack, a benchmark that evaluates agents on migrating ML repositories between these two ecosystems. Each agent receives a PyTorch repository and a clean TPU machine and must deliver a JAX repository. Its output passes if it meets our grading criteria, based on the source repository's published result, authenticated and calibrated by us. Migration by agents remains unreliable: Claude Opus 5 passes 75%, GPT-5.6-sol 45%, and every other agent at most 40%. XStack is a step toward freeing ML code from its ecosystem. We release XStack to make progress toward reliable migration across frameworks and hardware measurable.
Yanzhe Chen, Qifang Zhao, Xiaoxiao Xu, Fanghui Liu
Muon is increasingly used for language-model pretraining, yet its large-step dynamics are not captured by the classical edge-of-stability (EoS) picture of gradient descent (GD). In GD, loss neutrality, equal-magnitude update reversal, and marginal stability meet at a single learning-rate-dependent edge. We show that Muon breaks this coupling. For stochastic no-momentum Muon, we derive a coherence-corrected conditional loss-neutral boundary $2ρ_b/η$, while temporal alignment follows a separate geometry. Controlled experiments show that loss balance and temporal alignment respond differently to learning rate and batch size. Across our language model experiments, the 130M Llama-like LLM runs exhibit loss-boundary tracking with weak negative alignment, whereas the studied 1B LLM configuration shows stronger partial cancellation; in both settings, directions remain far from coherent reversal while training continues to improve. These results support a split EoS picture for Muon: a stochastic loss-neutral edge survives, but it is not accompanied by a universal temporal-direction signature. The source code for reproducing the experiments can be found in https://github.com/cyzebra/Muon-Sublates-the-Edge-of-Stability-in-LLM-Pretraining
Multimodal discrete diffusion language models (dLLMs) generate responses through iterative mask prediction and offer a promising alternative to autoregressive large vision-language models. However, they still suffer from hallucination, producing outputs that are plausible under language priors but insufficiently grounded in the image. We argue that hallucination in multimodal dLLMs is not only a token prediction problem, but also a premature commitment problem: during iterative denoising, some masked positions are unmasked too early because their confidence is driven more by language priors than by visual evidence, and these early commitments can bias later generation. To address this issue, we propose Visual Gain Ordering (VIGOR), a training-free and plug-and-play inference strategy that reorders masked positions according to counterfactual visual evidence. At each denoising step, VIGOR compares the confidence of each candidate token under the original image-conditioned pass and an additional visual-ablation pass, and prioritizes positions whose predictions are more strongly supported by the image. Experiments on LLaDA-V and Lumina-DiMOO show that VIGOR consistently reduces hallucination and improves multimodal reasoning performance across benchmarks. Code is available at https://github.com/THU-BPM/Vigor.
LLM-based RTL code generation methods increasingly target both functional correctness and PPA quality, yet existing approaches decouple the two objectives, optimizing PPA only after full correctness is achieved. Sequential multi-agent pipelines, evolutionary search with binary correctness gates, and hierarchical reward dependencies all discard partially correct but architecturally promising candidates. Moreover, existing methods reduce the multi-objective PPA space to a single scalar fitness that obscures the trade-offs among area, delay, and power. To address these limitations, we propose COEVO, a co-evolutionary framework that unifies correctness and PPA optimization within a single evolutionary loop. COEVO formulates correctness as a continuous co-optimization dimension alongside area, delay, and power, enabled by an enhanced testbench with fine-grained scoring. The framework allows PPA-promising but partially correct candidates to guide the search toward jointly optimal solutions, and further employs four-dimensional Pareto-based non-dominated sorting to preserve the PPA trade-off structure without manual weight tuning. On VerilogEval 2.0 and RTLLM 2.0, COEVO surpasses all agentic baselines across four LLM backbones. With GPT-5.4-mini, COEVO achieves 97.5\% and 94.5\% Pass@1 and attains the best PPA on 43 out of 49 synthesizable RTLLM 2.0 designs.
Dynamic point cloud pretraining is still dominated by masked reconstruction objectives. However, these objectives inherit two key limitations. Existing methods inject ground-truth tube centers as decoder positional embeddings, causing spatio-temporal positional leakage. Moreover, they supervise inter-frame motion with deterministic proxy targets that systematically discard distributional structure by collapsing multimodal trajectory uncertainty into conditional means. To address these limitations, we propose Diffusion Masked Pretraining (DiMP), a unified self-supervised framework for dynamic point clouds. DiMP introduces diffusion modeling into both positional inference and motion learning. It first applies forward diffusion noise only to masked tube centers, then predicts clean centers from visible spatio-temporal context. This removes positional leakage while preserving visible coordinates as clean temporal anchors. DiMP also reformulates point-wise inter-frame displacement supervision as a DDPM noise-prediction objective conditioned on decoded representations. This design drives the encoder to target the full conditional distribution of plausible motions under a variational surrogate, rather than collapsing to a single deterministic estimate. Extensive experiments demonstrate that DiMP consistently improves downstream accuracy over the backbone alone, with absolute gains of 11.21\% on offline action segmentation and 13.65\% under causally constrained online inference. Code is available at https://github.com/InitalZ/DiMP.
Existing attacks and defenses for coding agents act once content enters the model context: injected text persuades the model, and guards inspect what the model reads. Yet to judge whether processing an unknown archive, network response, or image is safe, an agent often must process it first, and that step can do harm before any content reaches the model. We call this first processing Step Zero, and harm done by it a Step-Zero attack. Across seven harnesses and six models, each agent received only a path or URL to a crafted input, paired with a benign twin showing the same visible fields. Step Zero is where most agents are already unsafe: 32/42 harness-model pairs overran a 4 MiB disk budget, 26/42 before the model saw any member, and four of the six models did so in every round (84/84), so the damage precedes any model-side defense. Every twin stayed within budget, so the attacker's bytes alone cause the harm. On network and image inputs the risk moves to what the agent writes: agents stored the body under the name the input declared, letting the input decide what is written, and their checks faded once cues were removed. Model-free tests trace the harm to the processing primitive, so we defend it there: a gate that counts the bytes each call produces held every routed extraction within budget while agents completed 25/28 tasks. Vendors should pair such gates with safety-alignment training that teaches agents to inspect an input before processing it.
AI 速览：AffordSim 将开放词汇 3D 可供性预测融入仿真轨迹生成，构建 50 任务基准，并验证生成数据可零样本迁移到真实 Franka FR3。
Many everyday robot manipulation skills are affordance-dependent, with success determined by whether the robot contacts the functional object region required by the subsequent action. Current simulation data generators obtain contacts from generic grasp estimators or per-object manual contact annotations, but generic estimators rank stable grasps without task semantics and often select contacts that are misaligned with the downstream action, while manual contact annotations must be rewritten for each new object and task. To solve these challenges, we introduce AffordSim, a scalable data generator and benchmark that integrates open-vocabulary 3D affordance prediction into simulation-based trajectory generation. Given a natural-language task description, AffordSim synthesizes a task-relevant scene, emits affordance queries, grounds them on object surfaces, samples region-conditioned grasps, and selects executable candidates with motion planning. It further randomizes object pose, texture, lighting, image noise, and cross-viewpoint backgrounds for sim-to-real transfer. We instantiate AffordSim as a 50-task benchmark across diverse manipulation skills, five robot embodiments, and 500+ rigid and articulated objects. AffordSim achieves 93% of the trajectory collection success rate of manual contact annotations on affordance-critical tasks and 89% on hard composite tasks. Vision-language-action policies trained on AffordSim data transfer zero-shot to a real Franka FR3, reaching 24% average success.
Video generation is shifting from end-to-end synthesis toward agentic generation, where planning, tool use, and feedback enable coherent, controllable multi-shot videos. In parallel, video world models provide action-conditioned observations and predicted futures, serving as interactive environments for agents. We study their intersection through the agent harness: an executable runtime that constructs context, exposes tools, maintains state, routes feedback, and manages budgets around video models. This report examines three directions. First, agentic video generation, where harnesses coordinate scripting, shot planning, asset management, production, and editing, alongside the evaluation challenges of these systems. Second, agentic world modeling, where harnesses mediate observation, control, execution, and state. We discuss hybrid designs in which game engines or agent-written programs maintain world state while video models render observations, and how agents and world models can mutually evaluate and improve one another. Third, recursive self-improvement (RSI) in video model development, distinguishing output revisions, updates to harness state, generator parameters, and evaluators, and changes to the improvement procedure itself. We identify how these improvements feed back into subsequent development. Across these directions, we analyze essential agent capabilities, including audiovisual perception, long-horizon planning, memory, code-based interaction, feedback-guided control, skill reuse, and multi-agent coordination. We conclude with challenges in autonomous video production, lifelong memory, programmable hybrid control, autonomous model development, provenance, cost accounting, and evaluation linking video quality and task performance to sustained improvements in models and harnesses.
Diffusion video models produce striking clips, but they cannot reliably render an exact formula, a correct chart, a rule-consistent game or narration synchronized to on-screen events. A different route is emerging: multimodal agents that write code to render every frame and synthesize every sound. We introduce VibeVideoBench, a benchmark for evaluating agents on code-driven video generation. Its 239 self-contained tasks in English and Chinese span 12 categories, from data stories and music-driven pieces to classroom lessons and paper explainers; each asks an agent, running in its native harness inside a sandbox with no image or video generation model, to deliver a finished MP4 with sound within one hour. Evaluation is hierarchical and evidence-based: program checks of the delivery spec, about 13k task-specific rubric checks traced to the requirements they enforce, and five quality dimensions anchored per task, scored by three VLM judge panels. Across nine frontier agent systems and 2,151 scored rollouts, GPT-6 Astra and Claude Opus 5.5 rank first and second with both the judges and human annotators, and the strongest system satisfies every core requirement on 76.3% of tasks. Delivery is a second axis: counting missing videos as zero changes the ranking, and three quarters of the lost videos are final renders that started too late, ran too slowly or never started, a failure of how agents plan their hour. Agents build their own rendering pipelines and inspect their own frames, yet how much they look does not predict quality. The judges' ranking of the systems closely matches blind human preferences.
Sizhe Chen, Sanjay Kariyappa, Liwei Jiang, Chong Xiang, Jihye Choi, Chaowei Xiao, G. Edward Suh
The recent incident in which OpenAI agents autonomously compromised Hugging Face's infrastructure illustrates the importance of instruction hierarchy (IH): an LLM should follow lower-priority instructions, such as user requests, only when they do not conflict with higher-priority instructions, such as system policies. IH can be effectively improved by fine-tuning LLMs on simulated examples of instruction conflicts, whose diversity is key to teaching LLMs to properly handle conflicts at test time. However, our decomposition of a task into goals and constraints reveals a conflict type -- goal-goal conflicts -- that is absent from existing IH training sets from OpenAI and NVIDIA. We show the importance of this conflict type: fine-tuning Qwen3.5-9B and Gemma4-12B-it on examples of this conflict type significantly improves IH on a wide range of real-world tasks unseen in training; adding these examples to existing IH training sets for fine-tuning also gives better IH adherence. The IH gains come with trivial mean utility changes on no-conflict tasks across eight benchmarks. Our work advocates including this new conflict type in IH training, even if these examples do not reflect typical LLM use.
Hongcheng Liu, Yuhao Wang, Zhe Chen, Pingjie Wang, Zhiyuan Zhu, Yixuan Hou Shanghai Jiaotong University, Yanfeng Wang Shanghai Jiao Tong University, Yu Wang Shanghai Jiao Tong University
Omni Large Language Models (Omni-LLMs) have demonstrated impressive capabilities in holistic multi-modal perception, yet they consistently falter in complex scenarios requiring synergistic omni-modal reasoning. Beyond understanding global multimodal context, effective reasoning also hinges on fine-grained cross-modal alignment, especially identifying shared referents across modalities, yet this aspect has been largely overlooked. To bridge this gap, we formalize the challenge as a cross-modal coreference problem, where a model must localize a referent in a source modality and re-identify it in a target modality. Building on this paradigm, we introduce CrossOmni, a dataset comprising nine tasks equipped with human-designed reasoning rationales to evaluate and enhance this capability. Experiments on 13 Omni-LLMs reveal systematic weaknesses in cross-modal coreference, which we attribute to the absence of coreference-aware thinking patterns. To address this, we enhance cross-modal alignment via two strategies: a training-free In-Context Learning method and a training-based SFT+GRPO framework designed to induce such thinking patterns. Both approaches yield substantial performance gains and generalize effectively to collaborative reasoning tasks. Overall, our findings highlight cross-modal coreference as a crucial missing piece for advancing robust omni-modal reasoning.
尚无 AI 解读，展示原文摘要
cs.CL · cs.CV
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Zhenhua Liu Soochow University, Lijun Li Shanghai Artificial Intelligence Laboratory, Ruizhe Chen, Yuxian Jiang, Tong Zhu Shanghai Artificial Intelligence Laboratory, Zhaochen Su, Wenliang Chen Soochow University, China, Jing Shao Shanghai AI Laboratory
While guided decoding, especially value-guided methods, has emerged as a cost-effective alternative for controlling language model outputs without re-training models, its effectiveness is limited by the accuracy of the value function. We identify that this inaccuracy stems from a core distributional gap: existing methods train static value functions on trajectories sampled exclusively from the base policy, which inherently confines their training to a narrow and suboptimal view of the potential output space. We propose Iterative Value Refinement, a novel framework designed to bridge this gap. It employs Value Exploration to provide a more comprehensive and robust training signal, complemented by Iterative Self-Refinement, which uses the improved value function from one iteration to guide the generation of higher-quality data for the next. Extensive experiments on text summarization, multi-turn dialogue, and instruction following demonstrate the effectiveness of our framework in aligning language models. Our approach not only achieves alignment but also significantly reduces computational costs by leveraging principled value function optimization for efficient and effective control.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Songtao Jiang, Yuan Wang, Ruizhe Chen, Yan Zhang ByteDance Inc., Ruilin Luo, Bohan Lei Zhejiang University, Yeying Jin, Sibo Song Alibaba Group 等 12 位作者
While reinforcement learning from verifiable rewards (RLVR) has been proven highly effective for enhancing reasoning, its application to medical visual question answering (Med-VQA) is hampered by models producing reasoning inconsistent with either the visual evidence or the final answer. Our analysis reveals a critical flaw in RLVR training: it paradoxically encourages models to disregard visual evidence and generate answers that contradict their own reasoning. This degradation is most pronounced in specialized medical modalities (e.g., Fundus, Ultrasound) where base VLMs lack robust understanding, a failure we attribute to a flawed reward mechanism exacerbated by the scarcity of diverse training data. To tackle this, we introduce Med-Zero-17K, a large-scale dataset spanning over 30 modalities and 24 clinically relevant tasks, and the Multi-Consistency Reward (MCR) framework, which explicitly rewards both perceptual grounding and logical coherence. Extensive experiments validate our approach: integrating MCR into the RLVR framework delivers robust performance gains. This success stems from our crucial finding that rewarding internal consistency is significantly more effective than attempting to judge reasoning correctness. Furthermore, MCR proves highly versatile, exhibiting strong generalization across diverse VLM backbones, compatibility with RL algorithms like GRPO and DPO, and extending its effectiveness to 3D VQA tasks and R1-style training paradigms. Code and dataset will be released.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Gemini Team, Petko Georgiev, Ving Ian Lei, Ryan Burnell, Libin Bai, Anmol Gulati, Garrett Tanzer, Damien Vincent 等 1137 位作者
In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content.