论文广场 - AcademicHub

01.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2606.15052

PANDA: An LLM-Enhanced Performance-Driven Analog Design Framework Bridging Design Intent and Layout Generation

作者:

Haoyi Zhang ↗Weijian Fan ↗Xiaohan Gao ↗Bingyang Liu ↗Runsheng Wang ↗Yibo Lin ↗

arXiv:2606.15052v1 Announce Type: cross Abstract: Traditional design of analog circuits heavily relies on manual interventions across topology, sizing, and layout, with prior automation addressing stages in isolation. In this work, we propose PANDA, an LLM-enhanced framework that bridges high-level design intent to final layout by actively managing cross-stage dependencies through guided topology synthesis, substructure-aware sizing, and constraint-driven layout generation. This shifts automation from algorithm-centric execution to intent-centric co-design, reducing turnaround time from days or weeks to hours while improving design performance.

阅读与讨论 → 访问原文 →

02.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.15007

Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning

作者:

We introduce Nemotron 3 Ultra, a 550 billion total and 55 billion active parameter Mixture-of-Experts Hybrid Mamba-Attention language model. We pre-trained Nemotron 3 Ultra on 20 trillion text tokens, then extended the context length to 1M tokens, and post-trained using Supervised Fine Tuning (SFT), Reinforcement Learning (RL), and Multi-teacher On-Policy Distillation (MOPD). Nemotron 3 Ultra is our most capable model yet, employing multiple key technologies - LatentMoE, Multi Token Prediction (MTP), NVFP4 pre-training, multi-environment RLVR, MOPD, and reasoning budget control. Nemotron 3 Ultra achieves up to ~6x higher inference throughput as compared to state-of-the-art publicly available LLMs while attaining on-par accuracy. The state-of-the-art accuracy, high inference throughput, and 1M token context length make Nemotron 3 Ultra ideal for long-running autonomous agentic tasks. We open-source the base, post-trained, and quantized checkpoints, along with the training data and recipe on HuggingFace.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.CL) 2026-06-11 DOI: arXiv:2606.12370

Breaking Entropy Bounds: Accelerating RL Training via MTP with Rejection Sampling

作者:

Yucheng Li ↗Huiqiang Jiang ↗Yang Xu ↗Jianxin Yang ↗Yi Zhang ↗Yizhong Cao ↗Yuhao Shen ↗Fan Zhou ↗Rui Men ↗Jianwei Zhang ↗An Yang ↗Bowen Yu ↗…

Reinforcement learning (RL) has become a key component in modern large language models, yet the rollout stage remains the key bottleneck in RL training pipelines. Although Multi-Token Prediction (MTP) offers a natural solution to accelerate rollouts through speculative decoding, many studies have observed that MTP acceptance rates degrade significantly during RL training, leading to limited speedup performance. To address this bottleneck, we present Bebop, a systematic study of MTP in LLM post-training, and offer practical recipes to integrate MTP into large-scale RL pipelines. First, we reveal that the MTP acceptance rate is fundamentally bounded by the fluctuation of model entropy, which demonstrates a clear negative linear relationship with the rise of entropy in the RL stage. Second, we show that probabilistic rejection sampling largely alleviates the disturbance introduced by entropy in RL compared to greedy draft sampling. We further identify that the conventional MTP training objectives (cross-entropy or KL) are suboptimal in such settings, and therefore we propose a novel end-to-end TV loss that directly optimizes multi-step rejection sampling acceptance rate, yielding ~10% acceptance rate improvements, achieving up to 95% acceptance rates and up to 25% extra inference throughput gains across mathematical reasoning, code generation, and agentic tasks. Third, we test various online MTP training strategies during RL and show that pre-RL MTP training with e2e TV loss and rejection sampling achieves a consistent acceptance rate and speedup throughout the entire RL, eliminating the need for costly online MTP updating. We provide extensive experiments and analysis that validate our findings. Experimental results show our method achieves up to 1.8x end-to-end acceleration in async RL training of Qwen3.5, Qwen3.6, and Qwen3.7 models.

阅读与讨论 → 访问原文 →

04.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2512.22560

RollArt: Disaggregated Multi-Task Agentic RL Training at Scale

作者:

Wei Gao ↗Yuheng Zhao ↗Tianyuan Wu ↗Shaopan Xiong ↗Weixun Wang ↗Dakai An ↗Lunxi Cao ↗Dilxat Muhtar ↗Zichen Liu ↗Haizhou Zhao ↗Ju Huang ↗Siran Yang ↗…

arXiv:2512.22560v2 Announce Type: replace-cross Abstract: Agentic Reinforcement Learning (RL) trains LLMs through multi-turn interactions with environments, producing workloads that mix compute-bound prefill, bandwidth-bound decoding, CPU-heavy environment execution, and bursty reward evaluation. Existing systems either colocate all stages on a single GPU cluster or decouple them only at a coarse granularity, overlooking hardware heterogeneity and incurring substantial synchronization overhead across stages. We present ROLLART, a system for multi-task agentic RL on disaggregated infrastructure. ROLLART maps each pipeline stage to best-fit hardware, routing prefill-heavy tasks to compute-optimized GPUs, decode-heavy tasks to bandwidth-optimized GPUs, and environments to CPU clusters. It decouples rollout at the trajectory level, allowing generation, environment interaction, and reward scoring to proceed independently, so that slow or failed environments never block the others. ROLLART offloads stateless reward computation to serverless infrastructure and overlaps rollout with training via staleness-bounded asynchronous weight synchronization. Our results demonstrate that ROLLART effectively improves training throughput and achieves 1.31–2.05 $\times$ training time reduction compared to various RL systems. We also evaluated ROLLART by training a hundreds-of-billions-parameter MoE model for Qoder product on an Alibaba cluster with above 3,000 GPUs, demonstrating its stability and scalability.

阅读与讨论 → 访问原文 →

05.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.16409

PathRouter: Aligning Rewards with Retrieval Quality in Agentic Graph Retrieval-Augmented Generation

作者:

Bo Wang ↗Heyan Huang ↗Yaolin Li ↗Wei Tang ↗Yuan Zhang ↗Wenbo Li ↗Mingze Gao ↗Ge Shi ↗Chong Feng ↗

Agentic GraphRAG trains language-model agents to iteratively retrieve and reason over graph-structured evidence, enabling more accurate and context-aware decision-making by efficiently navigating complex information networks. However, outcome-only reinforcement learning suffers from answer-path reward aliasing, where correct answers may come from shortcuts rather than useful evidence paths. It also exhibits search-update ambiguity, as scalar trajectory-level feedback does not indicate which retrieval actions to adjust. To mitigate these shortcomings, we present PathRouter, a path-aware training framework for agentic GraphRAG. PathRouter jointly evaluates each trajectory along answer correctness and evidence-path overlap, yielding four trajectory categories with differentiated GRPO advantage scaling that suppresses shortcut reinforcement while preserving evidence-seeking behavior. For evidence-poor trajectories, a frozen gold-evidence teacher provides token-level KL guidance on reasoning and search-query tokens, excluding answer tokens to avoid direct response imitation. Experiments on six QA benchmarks across three model sizes show that PathRouter consistently improves answer F1 and evidence-path overlap, achieving average F1 gains of 3.1 on 3B and 4.9 on 7B models compared to a strong baseline.

阅读与讨论 → 访问原文 →

06.

arXiv (CS.AI) 2026-06-11 DOI: arXiv:2606.11409

Risk Under Pressure: Compute-Aware Evaluation of Adversarial Robustness in Language Models

作者:

Malikeh Ehghaghi ↗Bogl\'arka Ecsedi ↗Marsha Chechik ↗Colin Raffel ↗

arXiv:2606.11409v1 Announce Type: cross Abstract: Adversarial robustness evaluations of large language models (LLMs) typically report attack success rate (ASR) under fixed query budgets, implicitly treating all attacks as equally costly. In practice, the computational expense of different attack strategies can vary by orders of magnitude. Consequently, ASR at a fixed budget can obscure the true effort required to jailbreak a model, thereby making it hard to determine whether an attack's cost justifies its payoff to the attacker. We propose a compute-aware evaluation framework based on computational pressure, measured in cumulative floating-point operations (FLOPs), as a proxy for adversarial effort. We introduce risk-compute curves, which map compute budgets to attack risk, and derive two metrics that summarize the average pressure required for a given attack to succeed. Across ten models spanning three families and four different stages in language model training and alignment, evaluated with three attack strategies (gradient-based, iterative refinement, and template-based) on two jailbreak robustness benchmarks, we find: (1) alignment training has non-monotonic effects on compute-space robustness; (2) scaling model size reduces gradient-based attack effectiveness but has limited impact on cheaper template-based attacks; (3) gradient-based attacks optimized on a surrogate model can transfer to a separate target model, providing a way to reduce attacker costs; (4) compute cost varies by up to ${\approx}5{\times}$ across harm categories within a single model; and (5) safety-aligned RL increases aggregate cost while leaving some categories disproportionately accessible. We release our framework to enable compute-aware risk assessment and evaluation.

阅读与讨论 → 访问原文 →

07.

arXiv (CS.CV) 2026-06-11 DOI: arXiv:2604.24119

TopoHR: Hierarchical Centerline Representation for Cyclic Topology Reasoning in Driving Scenes with Point-to-Instance Relations

作者:

Yifeng Bai ↗Zhirong Chen ↗Bo Song ↗Erkang Cheng ↗Haibin Ling ↗

Topology reasoning is crucial for autonomous driving. Current methods primarily focus on instance-level learning for centerline detection, followed by a sequential module for topology reasoning that relies on simplified MLP layers. Moreover, they often neglect the importance of point-to-instance (P2I) relationships in topology reasoning. To address these limitations, we present TopoHR (Topological Hierarchical Representation), a novel end-to-end framework that establishes cyclic interaction between centerline detection and topology reasoning, allowing them to iteratively enhance each other. Specifically, we introduce a hierarchical centerline representation including point queries, instance queries, and semantic representations. These multi-level features are seamlessly integrated and fused within a hierarchical centerline decoder. Furthermore, we design a hierarchical topology reasoning module that captures both fine-grained P2I relationships and global instance-to-instance (I2I) connections within a unified architecture. With these novel components, TopoHR ensures accurate and robust topology reasoning. On the OpenLane-V2 benchmark, TopoHR refreshes state-of-the-art performance with significant improvements. Notably, compared with previous best results, TopoHR achieves +3.8 in $\mathrm{DET}_{l}$, +5.4 in $\mathrm{TOP}_{ll}$ on $subset_A$ and +11.0 in $\mathrm{DET}_{l}$, +7.9 in $\mathrm{TOP}_{ll}$ on $subset_B$, validating the effectiveness of the proposed components. The code will be shared publicly at https://github.com/Yifeng-Bai/TopoHR.git.

阅读与讨论 → 访问原文 →

08.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2606.17648

From Brewing to Resolution: Tracing the Internal Lifecycle of Code Reasoning in LLMs

作者:

Siyue Chen ↗Yifu Guo ↗Yuquan Lu ↗Zishan Xu ↗Jiaye Lin ↗Jianbo Lin ↗Siyu Zhang ↗Cheng Yang ↗Junxin Li ↗Yujia Li ↗Yu Huo ↗Ruixuan Wang ↗…

arXiv:2606.17648v1 Announce Type: new Abstract: Standard accuracy metrics cannot explain why LLMs handle variable tracking but fail on semantically equivalent loops. We study an internal lifecycle of code reasoning in which models first brew the answer, making it linearly recoverable many layers before it becomes self-decodable, and then diverge into one of four resolution outcomes: Resolved, Overprocessed, Misresolved, or Unresolved. Understanding this lifecycle matters because similar task accuracies can mask fundamentally different failure modes that surface-level evaluation cannot detect. We introduce a dual diagnostic framework pairing layer-wise linear probing with Context-Stripped Decoding (CSD) and apply it to six code-reasoning task families across 16 models spanning Qwen, Llama, and DeepSeek architectures. All four outcomes carry substantial mass in every task family: overall Resolved is only 41.5%, with multiple tasks below 30%. Controlled sweeps over structure, depth, and operators expose task-specific failure bottlenecks: Function Call Resolved plunges from 61.1% to 2.5% as call depth increases from one to three. Across architectures and scales, the brewing scaffold remains stable, with normalized brewing duration 24-42% across all 16 models, while resolution success varies with capability. This indicates that the scaffold is a stable empirical regularity across the tested decoder-only Transformer families, whereas resolution success covaries with capability, scale, and training. Code: https://github.com/euyis1019/llm-brewing

阅读与讨论 → 访问原文 →

09.

arXiv (CS.CL) 2026-06-15 DOI: arXiv:2606.13931

DLawBench: Evaluating LLMs Through Multi-Turn Legal Consultation

作者:

Li Zhang ↗Yuzhen Shi ↗Yiran Hu ↗Jingwen Zhang ↗Wenbo Lv ↗Yubo Ma ↗Wei Wang ↗Rongyao Shi ↗Yuanyang Qiu ↗Xinran Xu ↗Yuemeng Qi ↗Linlin Miao ↗…

Lawyer-client consultation is a critical starting point for legal services. Effective legal assistance hinges on eliciting sufficient and truthful information from clients in order to devise strategies that best protect their interests. This task requires Large Language Models (LLMs) not only to perform robust legal reasoning, but also to strategically elicit material facts through multi-turn interactions and effectively guide clients with diverse personalities. Yet existing legal benchmarks overlook this interactive capability. To fill this gap, we introduce DLawBench, a diagnostic benchmark for real-world legal consultation. Drawing on realistic client behavior, we characterize lawyer-client interactions into four types: Cooperative, Dependent, Withdrawn, and Adversarial. Using dialogues grounded in real cases, DLawBench evaluates whether LLMs can effectively conduct legal consultation under realistic conditions. DLawBench comprises 461 cases from Chinese and U.S. law, 5,532 paired fact entries, 3,411 inquiry rubrics, and 3,348 issue-resolution rubrics, and evaluates 26 representative LLMs. Systematic experiments show substantial headroom: the best-performing model, GPT-5.5, achieves only 0.562 on consultation-grounded legal reasoning. More importantly, DLawBench exposes both sycophancy in legal consultation and a paradox: models perform worse when clients need guidance most.

阅读与讨论 → 访问原文 →

10.

arXiv (CS.CL) 2026-06-19 DOI: arXiv:2606.20002

Connect the Dots: Training LLMs for Long-Lifecycle Agents with Cross-Domain Generalization Via Reinforcement Learning

作者:

Yanxi Chen ↗Weijie Shi ↗Yuexiang Xie ↗Boyi Hu ↗Yaliang Li ↗Bolin Ding ↗Jingren Zhou ↗

This work presents a general framework for training large language models (LLMs) to "Connect the Dots" (CoD), a meta-capability required by long-lifecycle agents: as an LLM-based AI agent gets deployed in an environment, it solves a long sequence of tasks while continuously exploring the environment, learning from its own experiences, and iteratively self-updating its context about the environment, thereby achieving progressively better performance on future tasks conditioned on the updated context. Major components of the CoD framework include: (1) algorithm design and infrastructure for end-to-end reinforcement learning (RL) with long rollout sequences interleaving solve-task and update-context episodes; (2) tasks and environments for incentivizing and eliciting the targeted meta-capability in LLMs during training, as well as for faithfully measuring progress during evaluation. We present proof-of-concept implementations of the CoD framework, including a GRPO-style RL algorithm with fine-grained credit assignment, as well as tasks and environments tailored to the targeted meta-capability (rather than domain-specific LLM capabilities or standard task-by-task RL). Empirical results validate the efficacy of end-to-end RL training in the CoD setting, and demonstrate the potential for out-of-distribution generalization – within the training domains, across different domains, and from CoD to Ralph-loop settings – of the elicited meta-capability. Our investigation of CoD connects several lines of prior works, and opens up new opportunities for advancing LLMs and AI agents. To facilitate further research and applications, we release our implementations at \url{https://github.com/agentscope-ai/Trinity-RFT/tree/research/cod/examples/research_cod}.

阅读与讨论 → 访问原文 →

11.

arXiv (CS.LG) 2026-06-16 DOI: arXiv:2606.15637

HAPI-EP: Towards Hybrid, Adaptive, and Predictive Digital Twins of Cardiac Electrophysiology

作者:

Sumeet Vadhavkar ↗Xiajun Jiang ↗Yubo Ye ↗Maryam Toloubidokhti ↗Linwei Wang ↗

arXiv:2606.15637v1 Announce Type: new Abstract: A digital twin (DT) of a patient-specific heart offers significant potential in personalized medicine. However, its rapid and dynamic adaptation to an individual's live data and its predictive capability after adaptation remains central challenges. We examine this challenge from its two building blocks: DT formulation where mechanistic and data-driven models show competing merits and limitations, and DT optimization strategies that are largely driven by a reconstruction objective leading to un-identifiable models. We address both bottlenecks via HAPI – an AI framework for building hybrid, adaptive, and predictive DTs with three key enablers. First, HAPI constructs a physics-integrated gray-box model in which an interpretable mechanistic backbone is augmented by a neural component that models its residual to the observed data. Second, rather than attempting to pre-encode all possible variations in a static hybrid model, HAPI enables rapid on-the-fly adaptation of the hybrid model to few-shot live data, achieved by feedforward meta-learners realizing amortized inference of both mechanistic and neural parameters of the hybrid model trained with predictive objectives. Finally, we show that this adaptivity corresponds to the construction of a conditional generative model (i.e., the hybrid DT) that endows it with theoretical identifiability and thus strong performance in predictive scenarios. We demonstrate the proof-of-concept of HAPI in cardiac electrophysiology using a hybrid monodomain model with mechanistic reaction kinetics and neural graph diffusion. Across synthetic and real-data studies, we show that HAPI's mechanistic-neural hybridization and predictive adaptation are critical for obtaining identifiable DTs with strong predictive and out-of-distribution capabilities.

阅读与讨论 → 访问原文 →

12.

arXiv (CS.CV) 2026-06-17 DOI: arXiv:2507.04704

SPATIA: Multimodal Generation and Prediction of Spatial Cell Phenotypes

作者:

Zhenglun Kong ↗Mufan Qiu ↗John Boesen ↗Xiang Lin ↗Sukwon Yun ↗Tianlong Chen ↗Manolis Kellis ↗Marinka Zitnik ↗

Understanding how cellular morphology, gene expression, and spatial context jointly shape tissue function is a central challenge in biology. Image-based spatial transcriptomics technologies now provide high-resolution measurements of cell images and gene expression profiles, but existing methods typically analyze these modalities in isolation or at limited resolution. We address the problem by introducing SPATIA, a multi-level generative and predictive model that learns unified, spatially aware representations by fusing morphology, gene expression, and spatial context from the cell to the tissue level. SPATIA also incorporates a spatially conditioned generative framework with confidence-aware OT reweighting and morphology-profile alignment for modeling target-state morphology distributions. Specifically, we propose a confidence-aware flow matching objective that reweights weak optimal-transport pairs based on uncertainty. We further apply morphology-profile alignment to encourage biologically meaningful image generation, enabling the modeling of microenvironment-dependent phenotypic transitions. We assembled a multi-scale dataset consisting of 25.9 million cell-gene pairs across 17 tissues. We benchmark SPATIA against 18 models across 12 tasks, spanning categories such as phenotype generation, annotation, clustering, gene imputation, and cross-modal prediction. SPATIA achieves improved performance over state-of-the-art models, improving generative fidelity by 8% and predictive accuracy by up to 3%.

阅读与讨论 → 访问原文 →

13.

arXiv (CS.CL) 2026-06-12 DOI: arXiv:2606.12897

SafeLLM: Extraction as a Hallucination-Resistant Alternative to Rewriting in Safety-Critical Settings

作者:

Julia Ive ↗Felix Jozsa ↗Evridiki Georgaki ↗Nabeel Sheikh ↗Emma Cattell ↗Nick Jackson ↗Paulina Bondaronek ↗Ciaran Scott Hill ↗Richard Dobson ↗

Large language models (LLMs) are increasingly used to access organisational documentation, including standard operating procedures (SOPs), HR policies and institutional guidelines. However, retrieval-augmented generation (RAG) systems that rely on free-form rewriting can introduce hallucinations and unstable trade-offs between completeness and conciseness, particularly in safety- and compliance-critical settings. Objectives: To evaluate extraction as a hallucination-resistant alternative to rewriting-based RAG and compare strategies that balance precision, recall and safety across document types and model scales. Methods: We compare multiple prompting strategies, including line-number-based source selection, extraction of relevant guideline sentences with explicit safety annotations, and a multi-stage pipeline that refines draft answers using supporting evidence from source guidelines. Experiments are conducted on documents of varying length and structure, including local NHS acute care and oncology guidelines and UK-wide NICE guidelines, using both frontier-scale and locally deployable models. Performance is assessed using automatic metrics and human expert evaluation of relevance and completeness. Results: Line-number selection achieves the strongest results, outperforming direct copying and safety-focused strategies across both large and small models while maintaining high term recall (up to 95%) and close alignment with source text. Safety-oriented approaches improve precision but introduce systematic omissions, while multi-stage filtering further amplifies this trade-off. Performance varies with document structure: line-based extraction excels in protocol-like content, whereas alternative strategies perform better on more verbose documents (up to 97% term recall).

阅读与讨论 → 访问原文 →

14.

arXiv (CS.LG) 2026-06-18 DOI: arXiv:2511.05221

ActiTect: A Generalizable Machine Learning Pipeline for REM Sleep Behavior Disorder Screening through Standardized Actigraphy

作者:

arXiv:2511.05221v3 Announce Type: replace Abstract: Isolated rapid eye movement sleep behavior disorder (iRBD) is a major prodromal marker of $\alpha$-synucleinopathies, often preceding the clinical onset of Parkinson's disease, dementia with Lewy bodies, or multiple system atrophy. While wrist-worn actimeters hold significant potential for detecting RBD in large-scale screening efforts by capturing abnormal nocturnal movements, they become inoperable without a reliable and efficient analysis pipeline. This study presents ActiTect, a fully automated, open-source machine learning tool to identify RBD from actigraphy recordings. To ensure generalizability across heterogeneous acquisition settings, our pipeline includes robust preprocessing and automated sleep-wake detection to harmonize multi-device data and extract physiologically interpretable motion features characterizing activity patterns. Model development was conducted on a cohort of 78 individuals, yielding strong discrimination under nested cross-validation (AUROC = 0.95). Generalization was confirmed on a blinded local test set (n = 31, AUROC = 0.86) and on two independent external cohorts (n = 113, AUROC = 0.84; n = 57, AUROC = 0.94). To assess real-world robustness, leave-one-dataset-out cross-validation across the internal and external cohorts demonstrated consistent performance (AUROC range = 0.84-0.89). A complementary stability analysis showed that key predictive features remained reproducible across datasets, supporting the final pooled multi-center model as a robust pre-trained resource for broader deployment. By being open-source and easy to use, our tool promotes widespread adoption and facilitates independent validation and collaborative improvements, thereby advancing the field toward a unified and generalizable RBD detection model using wearable devices.

阅读与讨论 → 访问原文 →

15.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.04184

GroupToM-Bench: Benchmarking Group Theory of Mind and Nonlinear Social Emergence in MLLMs

作者:

Weidong Tang ↗Jierui Li ↗Yueling Hou ↗Zihan Mei ↗Can Zhang ↗Xinyan Wan ↗Zhiyuan Liang ↗Pengfei Zhou ↗Yang You ↗Wangbo Zhao ↗

True general intelligence requires not only a model of the physical world but also a social world model: the capacity to infer how individual mental states interact and crystallize into group-level outcomes. Despite notable progress in individual-level Theory of Mind (ToM) reasoning, existing multimodal large language models fail at this broader task. Collective behavior emerges non-linearly from social tensions, conformity dynamics, and structural constraints, meaning it cannot be recovered by merely summing individual intentions. We present GroupToM-Bench, the first multimodal benchmark for group-level ToM, built around a causal chain spanning micro-level BDI states (belief, desire, intention), meso-level group tension and structural constraints, and macro-level outcome prediction and mechanistic attribution. To probe this full arc, we develop a seven-level cognitive audit framework. Experiments reveal a gap between current models and human baselines, highlighting a failure to process social structures and non-linear collective dynamics.

阅读与讨论 → 访问原文 →

16.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.15932

Beyond NL2Code: A Structured Survey of Multimodal Code Intelligence

作者:

Xuanle Zhao ↗Qiushi Sun ↗Jingyu Xiao ↗Xuexin Liu ↗Haoyue Yang ↗Qiaosheng Chen ↗Xianzhen Luo ↗Jing Huang ↗Yufeng Zhong ↗Lei Chen ↗Shuai Fu ↗Zhenlin Wei ↗…

While LLMs have substantially advanced text-to-code synthesis, many real programming tasks specify intent through visual artifacts such as screenshots, charts, documents, vector drawings, videos, and interactive states. These tasks require models to connect visual perception to executable programs, because correctness depends not only on syntax but also on layout, geometry, data semantics, editability, interaction behavior, and domain-specific constraints that apply after execution. This survey examines Multimodal Code Intelligence, covering systems that generate, edit, refine, execute, or reason with code under visually grounded inputs and outputs. We first formulate the field by the role that code plays in each task, distinguishing code as a rendered artifact, an editable symbolic structure, a scientific representation, an intermediate reasoning trace, or an executable policy or tool interface. We then organize benchmarks and methods into four domains: Graphical User Interface, Scientific Visualization, Structured Graphics, and Frontier Tasks and Frameworks. This taxonomy connects mature artifact-generation problems to emerging agentic and unified settings and allows us to compare how different tasks treat evidence of correctness. Looking ahead, we argue that future research may benefit from four verification-centered directions. Multi-signal validation can combine complementary evidence of correctness, multi-state verification can test behavior across execution trajectories, cross-task transfer testing can probe reusable visual-code skills, and verifiable agent traces can reveal whether agent actions are grounded in visual evidence. Together, these directions may move multimodal code generation from single-output imitation toward evidence-grounded executable systems.

阅读与讨论 → 访问原文 →

17.

arXiv (CS.LG) 2026-06-12 DOI: arXiv:2606.12487

DynamicPTQ: Mitigating Activation Quantization Collapse via Residual-Stream Dynamics

作者:

Zimo Zhao ↗Maolin Wang ↗Bowen Yu ↗Bowen Liu ↗Xiao Han ↗Xiangyu Zhao ↗

arXiv:2606.12487v1 Announce Type: new Abstract: Post-training quantization (PTQ) is essential for efficient large language model inference, but reliably quantizing activations remains challenging when weights, activations, and KV caches are all quantized to 4-bit precision. A key difficulty lies in massive activations, whose extreme values dominate the activation range and amplify quantization errors. State-of-the-art methods mainly mitigate massive activations through transformation-based smoothing, such as orthogonal rotations and affine scaling, but overlook the cross-layer dynamics of the residual stream. In this paper, we show that massive activations emerge and disappear in a phase-wise pattern across network depth, triggering large residual changes. These changes cause newly injected layer-wise updates to dominate the 4-bit quantization scale and weaken historical residual information. To characterize this behavior, we introduce Jump Ratio and Historical Feature SNR. This suggests that static transformation-based smoothing cannot fully resolve dynamic quantization instability caused by cross-layer residual changes. Based on this analysis, we propose DynamicPTQ, a Dynamic Post-Training Quantization policy for phase-aware mixed-precision activation quantization. DynamicPTQ identifies quantization-sensitive layers from residual-stream dynamics and assigns 8-bit activation precision only to these layers, while keeping weights, KV caches, and other activations in 4-bit precision. It can be directly integrated with strong PTQ baselines such as QuaRot, SpinQuant, and FlatQuant. Experiments on LLaMA-2 and LLaMA-3 show that DynamicPTQ consistently improves perplexity and zero-shot QA performance under W4A4KV4 quantization, while achieving 1.05 to 1.07 times throughput improvement with modest memory overhead. These results demonstrate a practical path toward robust low-bit LLM inference.

阅读与讨论 → 访问原文 →

18.

arXiv (CS.CL) 2026-06-19 DOI: arXiv:2606.19625

Where Does Social Reasoning Come From? Capability Provenance in Language Models

作者:

Glenn Matlin ↗Chandreyi Chakraborty ↗Saehee Eom ↗Mika Okamoto ↗Rayan Castilla ↗Louis Jaburi ↗Alvin Deng ↗Taywon Min ↗Lucia Quirke ↗Stella Biderman ↗Mark Riedl ↗

We use training-data attribution as an interpretable tool for capability discovery, mapping which regions of the pretraining corpus support social-reasoning versus STEM-reasoning in OLMo3-7B. Training-data attribution measures how strongly each training document influences a model's predictions on a benchmark, but document-level scores are too noisy to identify which corpus regions support which capabilities, and prior work has emphasized factual knowledge rather than reasoning. We compute gradient-based attribution (TrackStar via Bergson) over a working set drawn from the de-duplicated Dolma3 mix, aggregate influence across WebOrganizer's 24-format x 24-topic taxonomy (576 bins), and contrast benchmark pairs in a 2x2 design that varies domain (social vs. STEM) and capability type (reasoning vs. knowledge): SocialIQA and MMLU Social Sciences against ARC-Challenge and MMLU STEM. Social and STEM reasoning draw on qualitatively distinct corpus regions, and the contrast is sharper at the reasoning level than at the knowledge level. Targeted machine unlearning provides partial causal validation: forgetting high-attribution topic bins (e.g., Literature for SocialIQA) degrades the aligned benchmark more than within-bin random baselines, and we open-source all code, sampling manifests, the bin-level influence matrix, and unlearning checkpoints.

阅读与讨论 → 访问原文 →

19.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2604.18827

OmniMouse: Scaling properties of multi-modal, multi-task Brain Models on 150B Neural Tokens

作者:

arXiv:2604.18827v2 Announce Type: replace-cross Abstract: Scaling data and artificial neural networks has transformed AI, driving breakthroughs in language and vision. Whether similar principles apply to modeling brain activity remains unclear. Here we leveraged a dataset of 3.1 million neurons from the visual cortex of 73 mice across 323 sessions, totaling more than 150 billion neural tokens recorded during natural movies, images and parametric stimuli, and behavior. We train multi-modal, multi-task models that support three regimes flexibly at test time: neural prediction, behavioral decoding, neural forecasting, or any combination of the three. OmniMouse achieves state-of-the-art performance, outperforming specialized baselines across nearly all evaluation regimes. We find that performance scales reliably with more data, but gains from increasing model size saturate. This inverts the standard AI scaling story: in language and computer vision, massive datasets make parameter scaling the primary driver of progress, whereas in brain modeling – even in the mouse visual cortex, a relatively simple system – models remain data-limited despite vast recordings. The observation of systematic scaling raises the possibility of phase transitions in neural modeling, where larger and richer datasets might unlock qualitatively new capabilities, paralleling the emergent properties seen in large language models. Code available at https://github.com/enigma-brain/omnimouse.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.LG) 2026-06-15 DOI: arXiv:2506.05797

EqCollide: Equivariant and Collision-Aware Deformable Objects Neural Simulator

作者:

Qianyi Chen ↗Tianrun Gao ↗Chenbo Jiang ↗Tailin Wu ↗

arXiv:2506.05797v2 Announce Type: replace Abstract: Simulating collisions of deformable objects is a fundamental yet challenging task due to the complexity of modeling solid mechanics and multi-body interactions. Existing data-driven methods often suffer from lack of equivariance to physical symmetries, inadequate handling of collisions, and limited scalability. Here we introduce \name, the first end-to-end equivariant neural fields simulator for deformable objects and their collisions. We propose an equivariant encoder to map object geometry and velocity into latent control points. A subsequent equivariant Graph Neural Network-based Neural Ordinary Differential Equation models the interactions among control points via collision-aware message passing. To reconstruct velocity fields, we query a neural field conditioned on control point features, enabling continuous and resolution-independent motion predictions. Experimental results on 2D and 3D scenarios show that \name achieves accurate, stable, and scalable simulations across diverse object configurations. It achieves $24.34\%$ to $57.62\%$ lower rollout MSE, even compared with the best-performing baseline model. Furthermore, \name could generalize to more colliding objects and extended temporal horizons, and stay robust to input transformed with group action. Code is available at: https://github.com/AI4Science-WestlakeU/EqCollide

阅读与讨论 → 访问原文 →

21.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2605.21028

DySink: Dynamic Frame Sinks for Autoregressive Long Video Generation

作者:

Bo Ye ↗Xinyu Cui ↗Jian Zhao ↗Tong Wei ↗Min-Ling Zhang ↗

Autoregressive long video generation often adopts bounded-memory streaming for efficiency, typically combining local windows for short-term continuity with static early-frame sinks as long-range anchors. However, this fixed allocation keeps early frames cached even when the current visual state has substantially diverged from them, while discarding potentially more relevant intermediate history. As a result, the retained long-range context may become less adaptive and bias generation toward outdated cues; in severe cases, RoPE-induced phase re-alignment can homogenize inter-head attention and cause sink collapse, where content regresses toward sink frames. We propose DySink, a retrieval-based framework that maintains a compact memory bank and selects visually relevant historical frames as dynamic frame sinks. DySink couples adaptive retrieval with a sink anomaly gate, which detects excessive inter-head consensus over retrieved context and suppresses collapse-prone context. Experiments on minute-long videos show that DySink consistently improves dynamic degree over strong baselines while also achieving higher temporal quality. The code and model weights will be released at https://github.com/yebo0216best/DySink.

阅读与讨论 → 访问原文 →

22.

arXiv (quant-ph) 2026-06-19 DOI: arXiv:2606.20262

Anomalous magneto-optical response at $\mathrm{RuO_2 / WSe_2}$ van der Waals interface

作者:

arXiv:2606.20262v1 Announce Type: cross Abstract: Ruthenium dioxide ($\mathrm{RuO_2}$) has been proposed as an altermagnetic candidate, although its magnetic ground state remains controversial. Here, we probe weak interfacial magnetic states at the surface of (001)-oriented $\mathrm{RuO_2}$ films using the magnetic proximity effect (MPE) in a van der Waals heterostructure consisting of monolayer tungsten diselenide ($\mathrm{WSe_2}$) atop $\mathrm{RuO_2}$. Temperature-dependent magneto-optical spectroscopy reveals an anomalous excitonic energy shift and a deviation from conventional Varshni behavior below 55 K that are absent in an encapsulated $\mathrm{WSe_2}$ control sample. The anomalous shift reverses sign upon field cooling with opposite magnetic field polarity, indicating a magnetic origin. Polarization-resolved measurements further show a nearly field-independent and fluctuating valley splitting in $\mathrm{WSe_2 / RuO_2}$ in strong contrast to the conventional linear Zeeman splitting observed in the control bare $\mathrm{WSe_2}$ sample. These results suggest that the valley states are governed predominantly by interfacial exchange fields associated with weak surface magnetic states in $\mathrm{RuO_2}$, which do not produce a conventional linear Zeeman response within the applied magnetic field range. Importantly, this approach enables direct optical probing of emergent surface magnetism without introducing an additional ferromagnetic layer, positioning MPE-based optical probing as a tool for investigating weak surface magnetism and offering new possibilities for studying magnetic materials with controversial magnetic states.

阅读与讨论 → 访问原文 →

23.

arXiv (CS.LG) 2026-06-17 DOI: arXiv:2602.03045

Clarify Before You Draw: Proactive Agents for Robust Text-to-CAD Generation

作者:

Bo Yuan ↗Zelin Zhao ↗Petr Molodyk ↗Bin Hu ↗Yongxin Chen ↗

arXiv:2602.03045v2 Announce Type: replace Abstract: Large language models have recently enabled text-to-CAD systems that synthesize parametric CAD programs (e.g., CadQuery) from natural-language prompts. In practice, however, geometric descriptions can be under-specified or internally inconsistent: critical dimensions may be missing and constraints may conflict. However, existing fine-tuned models tend to reactively follow the user instructions and hallucinate dimensions when the text is ambiguous. To address this, we propose a proactive agentic framework for text-to-CadQuery generation, named as ProCAD, that resolves specification issues before code synthesis. Our framework pairs a proactive clarifying agent, which audits the prompt and asks targeted clarification questions only when necessary to produce a self-consistent specification, with a CAD coding agent that translates the specification into an executable CadQuery program. We fine-tune the coding agent based on a curated high-quality text-to-CadQuery dataset and train the clarifying agent via agentic SFT on clarification trajectories. Experiments show that proactive clarification significantly improves robustness to ambiguous prompts while keeping interaction overhead low. ProCAD outperforms frontier closed-source models, including Claude Sonnet 4.5, reducing the mean Chamfer distance by 79.9% and lowering the invalidity ratio from 4.8% to 0.9%. Our code and datasets are made publicly available on https://github.com/BoYuanVisionary/Pro-CAD.

阅读与讨论 → 访问原文 →

24.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.15880

Deep Residual Injection for Full-Spectrum Forensic Signal Perception in Multimodal Large Language Models

作者:

Kaiqing Lin ↗Zhiyuan Yan ↗Ruoxin Chen ↗Ke-Yue Zhang ↗Yue Zhou ↗Caiyong Piao ↗Bin Li ↗Taiping Yao ↗Bo Wang ↗Youchang Xiao ↗Shouhong Ding ↗

Multimodal large language models (MLLMs) have been increasingly adopted in forensics for their robust semantic understanding. As AI-generated images become realistic, semantic-level inconsistencies alone are often insufficient for reliable detection. This motivates a critical question: whether MLLMs can achieve full-spectrum forensic signal perception, i.e., capturing low-level generator artifacts without sacrificing pre-trained semantic knowledge. We further perform a layer-wise analysis of forensic signal perception in MLLMs, showing that semantic information is primarily formed in the early-to-middle layers, whereas direct fine-tuning for artifact learning disrupts these semantic representations. Based on this insight, we propose Deep Visual Residual MLLM (Deep-VRM) to preserve early semantic processing while injecting artifact-specific visual signals as a residual path into an intermediate layer, where they are fused with semantic token representations and propagated through subsequent trainable layers. This enables later layers to jointly model semantic reasoning and signal-level forensic cues, and surprisingly, the model learns to adaptively leverage different levels of forensic signals depending on the input, achieving robust and generalizable detection performance. Extensive experiments show that our method achieves state-of-the-art across most benchmarks. The code and data are available at https://github.com/KQL11/Deep-VRM.

阅读与讨论 → 访问原文 →

25.

arXiv (CS.AI) 2026-06-11 DOI: arXiv:2505.13196

A Physics-Inspired Optimizer: Velocity Regularized Adam

作者:

Pranav Vaidhyanathan ↗Lucas Schorling ↗Natalia Ares ↗Maike Osborne ↗

arXiv:2505.13196v3 Announce Type: replace-cross Abstract: We introduce Velocity-Regularized Adam (VRAdam), a physics-inspired optimizer for training deep neural networks that draws on ideas from quartic terms for kinetic energy with its stabilizing effects on various system dynamics. Previous algorithms, including the ubiquitous Adam, operate at the so-called adaptive edge of stability regime during training, leading to rapid oscillations and slowed convergence of loss. However, VRAdam adds a higher order penalty on the learning rate based on the velocity such that the algorithm automatically slows down whenever weight updates become large. In practice, we observe that the effective dynamic learning rate shrinks in high-velocity regimes, and damping oscillations. By combining this velocity-based regularizer for global damping with per-parameter scaling of Adam, we create a powerful hybrid optimizer. For this optimizer, we provide rigorous theoretical analysis of operation at the edge of stability from a physical and control perspective for the momentum. Furthermore, we derive convergence bounds with the rate $\mathcal{O}(\ln(N)/\sqrt{N})$ for a stochastic non convex objective under mild assumptions. We demonstrate that VRAdam exceeds the performance against standard optimizers including AdamW. We benchmark various tasks such as image classification, language modeling, and generative modeling using diverse architectures and training methodologies including Convolutional Neural Networks (CNNs), Transformers, and GFlowNets.

阅读与讨论 → 访问原文 →

探索全球前沿学术脉络