LucidLyra: Feed-Forward Generative 3D Scene Reconstruction and Joint Understanding via Multi-Task Distillation
Oliver Hahn, Rodrigo Marcuzzi, Sangeetha Grama Srinivasan, Alessandro Burzio, Sherwin Bahmani, Jiahui Huang, Stefan Roth, Shengyu Huang 等 10 位作者
AI 速览：LucidLyra 从视频潜变量出发，以前馈方式生成可查询语义与实例的三维高斯场景，无需逐场景优化。
We present LucidLyra, a framework that makes generatively reconstructed 3D scenes directly queryable for semantics and instances in a feed-forward manner from video latents. While generative 3D reconstruction can synthesize virtual environments from a single image, downstream applications in robotics and physical AI require not only visual realism but also access to scene semantics and object identities. Existing 3D understanding approaches, however, require captured multi-view scenes, have limited generalization capabilities, or rely on expensive per-scene processing and optimization. To bridge this gap, our method distills complementary knowledge from a camera-controlled video diffusion model and frozen 2D vision foundation models into a unified model to jointly infer 3D Gaussians encoding geometry, appearance, open-vocabulary semantics, and class-agnostic instance cues. The resulting scenes support novel-view rendering, open-vocabulary segmentation, and object grouping without per-scene optimization, while training requires neither manual 3D annotations nor captured multi-view data. Our method outperforms baselines combining state-of-the-art 3DGS segmentation approaches with generative 3D scene reconstruction on curated synthetic and real-image-conditioned benchmarks, while preserving reconstruction quality. Furthermore, our latent formulation naturally extends to monocular video, reconstructing structured dynamic 3D scenes across time. By turning renderable but unstructured outputs into queryable 3D worlds, LucidLyra enables downstream applications including 3D scene editing, 3D asset mining, and robotic simulation.
LLM-based assistants on web-scale service platforms increasingly address local-service needs through multi-step planning and tool use. Under dynamic business conditions and evolving user needs, outcome-level evaluation offers limited evidence of where execution fails or how service drifts across continued interaction. We present ATLAS, a dual-horizon diagnostic evaluation framework for industrial tool-use agents. Within-turn evaluation uses trajectory-wise signals across thinking and reflection, tool and skill execution, and response generation to identify execution failures and unmet capability requirements. Across-turn evaluation uses user-wise signals spanning multiple lifecycles to diagnose failures in maintaining or updating user intent, constraints, and corrections. Fine-grained signals use rule-based checks and LLM judgments with explicit evidence scopes and decision boundaries. To support reliable diagnosis at production scale, LLM judges are calibrated against dynamically maintained references from business logs, and selected judges are distilled into compact diagnostic models. The resulting diagnostic profiles help developers attribute failures, prioritize targeted interventions, and reassess improvements. Applied in Meituan Xiaotuan, a million-scale DAU local-services assistant, ATLAS achieves high diagnostic fidelity across calibrated judges and distilled models. Blinded human auditing and a 14-day online A/B test with nearly two million users per arm confirm that the ATLAS-guided agent improves recommendation relevance, factual reliability, user engagement (+2.84% answer read-through rate), and order volume (+6.13% paid orders), leading to full production deployment.
Online vectorized map construction provides autonomous vehicles with essential, real-time semantic and geometric scene understanding. While leveraging prior maps presents an opportunity to improve accuracy, they are often biased due to outdated information or temporary road changes. Existing methods, which typically treat priors as fully reliable or only consider limited noise patterns, suffer severe performance degradation when these priors are imperfect. To address this, we propose PriorMaskMap, a robust framework for prior-guided vectorized map construction. Its first innovation is a Prior-Aware Confidence Estimator (PACE) that evaluates the input rasterized map against current visual observations. It assigns confidence scores to regions classified as reliable, erroneous, or blank, creating a foundational mask that guides two subsequent branches. In one branch, the Dual-Prior Query Generator (DPQG) selectively filters and embeds both rasterized and vectorized priors to initialize informative queries from trustworthy map structures. In the other, the Confidence-Guided Prior Fusion (CGPF) module spatially propagates and refines the mask, using the refined version to govern the fusion of BEV features with the prior map, thereby providing rich contextual representations for the map decoder. Evaluation on the nuScenes dataset shows that PriorMaskMap achieves superior mapping precision over existing approaches and exhibits marked robustness when provided with priors containing apparent errors. Our code is released at https://github.com/healenrens/PriorMaskMap.
Rotation equivariance is an important structural prior for a broad range of vision tasks, particularly in domains such as remote sensing and medical imaging where objects have no canonical orientation, yet it remains notably absent from current Mamba-based vision architectures. Despite the success of Mamba in natural language processing and its growing adoption in computer vision, existing visual Mamba models fail to account for rotational symmetry in their design. This omission renders them sensitive to image rotations, thereby constraining their robustness and cross-task generalization. To address this limitation, we incorporate rotation symmetry, an important geometric prior in many visual domains, into Mamba-based architectures. Specifically, we introduce EQ-VMamba, the first rotation equivariant visual Mamba architecture for vision tasks. The core components of EQ-VMamba include a carefully designed rotation equivariant cross-scan strategy and group Mamba blocks. Moreover, we provide a rigorous theoretical analysis of the intrinsic equivariance error, demonstrating that the proposed architecture enforces end-to-end rotation equivariance under the p4 rotation group. Extensive experiments across multiple benchmarks---including high-level image classification, mid-level semantic segmentation, and low-level image super-resolution---demonstrate that EQ-VMamba consistently improves rotation robustness and achieves superior or competitive performance compared to non-equivariant baselines on most benchmarks, while requiring 40-70% fewer parameters. These results indicate that embedding rotation equivariance not only effectively bolsters the robustness of visual Mamba models against rotation transformations, but also improves accuracy on data exhibiting rotational symmetry, with substantially improved parameter efficiency. Code is available at https://github.com/zhongchenzhao/EQ-VMamba.
As AI accelerator architectures become increasingly diverse, integrating different architectures into a unified system is emerging as a new heterogeneous computing paradigm. GPUs and WSEs differ substantially in compute, memory, and communication organization, creating both architectural complementarity and additional overheads, yet how to translate this heterogeneity into practical performance gains remains unclear. To address this gap, we propose a hierarchical benchmarking methodology to characterize GPU--WSE heterogeneous execution and the conditions for exploiting complementarity, together with an analytical performance model that identifies when heterogeneity becomes a bottleneck and when it yields system-level performance gains.
Haoyu Hu, Xuandong Zhao, Xuhai "Orson'' Xu, Nori Jacoby
Reinforcement learning with verifiable rewards (RLVR) generates hundreds of thousands of tokens per training step, with rollout generation dominating the computational cost. The overall token budget can be controlled along two main dimensions: (i) deciding which prompts to allocate rollouts to, and (ii) deciding how long each rollout should be. Prior work has generally controlled only one of these dimensions at a time. We show that jointly tuning both decisions under a shared compute budget improves both reasoning quality and wall-clock training time. We instantiate this view as \textbf{DU}al-controlled tok\textbf{E}n alloca\textbf{T}ion (DUET), a computationally efficient layer over GRPO that uses a lightweight pre-rollout surrogate of prompt informativeness to set how many rollouts each prompt receives, and a marker-gated abort rule with importance reweighting to set when to stop them. On Qwen3-1.7B trained on MATH, DUET outperforms full-budget GRPO and the other three budget-aware baseline methods. DUET's advantage further generalizes to other benchmarks across math and coding, and is on par with the best baseline on the scientific Q\&A domain, while also achieving a $1.62\times$ wall-clock speedup. More notably, using only 50\% of the token budget, DUET still outperforms all baseline methods at their full budget, achieving an even higher $2.51\times$ speedup over full-budget GRPO. We verify the high performance of DUET on other backbone LLMs, including Qwen3-4B to 14B and Llama-3.2-3B-Instruct. Notably, the gap between DUET and the strongest baseline \emph{widens} as the budget tightens, contrary to the usual pattern in which efficient methods trade off quality as compute decreases. More broadly, these results suggest that DUET budget-aware control strategies are valuable not only for accelerating training, but also for improving the quality of the learning signal.
Ziwei Zhang, Tianyin Liao, Chunyu Hu, Ge Lan, Nuo Xu, Yi Huang, Qingyun Sun, Xingxuan Zhang 等 10 位作者
Graph Foundation Models (GFMs) have recently emerged as a promising paradigm for learning transferable representations across different graphs and domains. However, existing GFMs face notable limitations, including limited pre-training scale or reliance on aligning with textual information in Large Language Models (LLMs). To overcome these limitations, recent studies have drawn inspiration from prior-data fitted networks (PFNs), which learn from large synthetic datasets and support multi-domain prediction backed by causal learning. However, existing graph PFNs encounter severe difficulties in modeling causal relationships within and across node and edge sets, generating diverse graph structures, and designing suitable architectures that can learn from complex synthetic datasets. To address these challenges, we propose **GSCM-PFN** (Graph SCM-based Prior-Data Fitted Network), a graph PFN pre-trained on synthetic data generated from an SCM-based prior that jointly models nodes and graph topology. Specifically, we (1) design a joint feature-edge-label structural causal model (SCM) to guide synthetic graph generation and capture causal relationships within and across node and edge sets; (2) introduce prior-guided node-edge dependency modeling to instantiate diverse graph generation strategies, and prove that our SCM-based generation prior is more expressive than existing priors; and (3) propose a feature-permutation-invariant dual-path graph network with theoretical guarantees for learning from diverse synthetic graphs. Experiments on 23 real-world benchmarks show that GSCM-PFN outperforms both conventional GFMs and PFN-based GFMs, without utilizing real-world graph data for pre-training. Code: https://anonymous.4open.science/r/GSCM-PFN-A3A3/
Chenyi Li, Haoyu Hu, Raja Marjieh, Katherine M Collins, Max Kreminski, Ilia Sucholutsky, Dalton Conley, Nori Jacoby
As AI agents increasingly participate in creative teamwork, writing emerges from hybrid human-AI teams. Although AI can improve the creativity of individual outputs, these gains may come at the cost of homogenization. What is the best role for human intervention? We introduce a transmission-chain paradigm inspired by cognitive models of writing, with four stages: Plan, Write, Review, and Revise. Each stage receives the previous output and passes its own forward. Using data from 333 participants across 427 stories, we compared creativity, diversity, and enjoyment across human, AI, and hybrid chains. Consistent with prior work, all-AI chains produced more creative but less diverse stories than all-human chains. Introducing a single human into an otherwise all-AI chain increased diversity depending on role. Human planning provided the best balance: it preserved AI-level creativity, increased diversity that persisted through downstream AI stages, and maintained enjoyment. We discuss implications for designing human-in-the-loop creative pipelines.
We propose a multimodal transformer framework for efficiently modeling long1 financial documents that combine structured tabular data with noisy, unstructured2 textual disclosures. The framework represents financial tables as semantically3 enriched token sequences and integrates them with long-form text through a cross-4 attention mechanism that explicitly captures cross-modal dependencies. In par-5 ticular, we augment each numerical variable with a semantic embedding derived6 from its textual definition, enabling the model to encode both numerical magnitude7 and economic meaning. This design yields a more expressive and parameter-8 efficient representation of tabular data compared to value-only embeddings. Using9 a large-scale dataset of financial statements and MD&A disclosures, we show that10 structured numerical features remain the dominant source of predictive signal;11 however, textual information provides consistent incremental gains when incor-12 porated via cross-attention rather than simple concatenation. Our results further13 demonstrate that noisy textual disclosures are most useful when conditioned on a14 stable, semantically structured tabular representation. In particular, using frozen15 variable-level tabular representations as cross-attention queries enables the model16 to weight MD&A chunks differently for each financial variable and observation.17 These findings suggest that cross-attention provides an efficient mechanism for18 table-conditioned aggregation of long-form narrative evidence.
My Chiffon Nguyen, Tianyi Hu, Vinod Anbalagan, Henry Gagnier, Pouya Sadeghi, Farah Abdou, Tidor-Vlad Pricope, Diego MMS Oliveira 等 164 位作者
As LLMs expand their language coverage, evaluating how well they understand and represent diverse cultures becomes increasingly important. However, many benchmarks remain English-centric, with limited cultural coverage and a focus on factual knowledge over its application in real-world cultural contexts. We propose CulturalRiddles, a benchmark of original, culturally grounded riddles created entirely by humans through a participatory community effort, spanning 62 cultures (45 countries, 51 languages). These riddles are deeply rooted in their respective cultures and require abductive and compositional reasoning, alongside detailed factual and social knowledge, language skills, and an understanding of culturally specific associations. We benchmark 28 LLM configurations using both automatic and large-scale human evaluation. Even frontier models achieve at most 22% accuracy in some communities. We publicly release all artifacts alongside an open contribution process to expand the benchmark's linguistic and cultural coverage.
Zihan Fang, Ron Asoulin-Handelman, Helen Wu, Yu Huang, Gina R. Bai
Recent advances in generative AI have enabled assistants to generate increasingly sophisticated executable unit tests. However, novices may accept these outputs without independently identifying what to test, evaluating test-suite completeness, or verifying correctness, potentially limiting opportunities to develop essential testing skills. Although prior work has explored AI-assisted test generation and feedback-oriented programming tutors, little is known about whether AI can support novices' testing practices without generating tests for them. To investigate this, we introduce TestCompass, a generative AI tutor that provides targeted feedback on weaknesses in students' test suites without generating executable tests. In a classroom study with 18 students, we compared TestCompass with a general AI assistant across two Java unit-testing tasks, examining students' testing behaviors, test-suite quality, and subjective experiences. We found that participants using the general-purpose AI assistant produced more tests, whereas TestCompass participants spent more time developing and verifying tests themselves. However, most test-suite quality measures were comparable between conditions, and TestCompass participants achieved similar requirement coverage with significantly fewer test methods ($p=0.012$) in one task. Neither group reported short-term improvements in perceived testing knowledge or confidence. These findings suggest that feedback-oriented AI can support meaningful testing outcomes while preserving opportunities for hands-on testing practice, offering a complementary approach to AI-assisted testing education.
Reinforcement learning (RL) has shown promise for training multimodal agents to reason and use tools. However, it remains unclear when fine-grained credit assignment improves learning and when it fails to improve task performance. Video editing provides a useful testbed for this question, combining multimodal perception, long-horizon planning, and tool use with delayed outcome feedback. We construct four movie-trailer editing task types---Content Selection, Narrative Sequencing, Rough Cut, and Fine Cut---that reflect real-world editing workflows and span different levels of task complexity and interaction horizon. Decaying process rewards provides the largest benefit on Fine Cut, with a 45.4% relative gain over outcome-only GRPO on tasks averaging more than 60 turns under outcome-only GRPO. On the three shorter-horizon task types averaging 30--40 turns under the same baseline, outcome-only GRPO remains competitive, with decaying process rewards yielding at most a 10.4% relative gain. Process rewards from an LLM judge initially improve tool use and encourage visual inspection. As training progresses, however, GRPO+PRM with a fixed process-reward weight increasingly exhibits excessive reasoning and redundant tool calls: process rewards continue to rise while task completion rates and outcome scores decline. Decaying the process-reward weight to zero mitigates this reward-hacking behavior while preserving the early gains. Despite dense token-level supervision, on-policy self-distillation (OPSD) yields little improvement or substantial degradation and is highly sensitive to the choice of privileged information. Our analysis reveals differences in teacher--student disagreement and gradient magnitudes, and suggests possible dilution of informative signals over long trajectories.
Multi-Turn Long-Form Question Answering (MT-LFQA) is a key application paradigm of Large Language Models (LLMs) in knowledge-intensive domains. However, existing benchmarks are limited to single-turn dialogue, while multi-turn dialogue benchmarks typically assess other orthogonal capabilities rather than knowledge-intensive factuality. To bridge this critical gap, we introduce \textbf{KnowMT-Bench}, a new benchmark designed to systematically evaluate MT-LFQA for LLMs across knowledge-intensive fields, including medicine, finance, and law. To faithfully assess the model's real-world performance, KnowMT-Bench employs a dynamic evaluation setting where models generate their own multi-turn dialogue histories given logically progressive question sequences. Our experiments on a diverse suite of LLMs show a clear degradation in both factual capability and information delivery efficiency in multi-turn dialogue. We further probe the underlying causes and investigate mitigation strategies spanning domain-specific finetuning, active context denoising, and RAG. The results show domain-dependent gains from finetuning, while active context denoising and RAG more consistently mitigate these issues, with RAG notably reversing this performance degradation. These findings underscore the importance of our benchmark for evaluating LLMs' conversational factual capabilities in real-world applications. Code and data are publicly available.\footnote{\href{https://github.com/Sven-LI-sankyuu/KnowMT-Bench}{\texttt{https://github.com/Sven-LI-sankyuu/KnowMT-Bench}}}
Fan Xu, Xinyu Hu, Zhenghan Yu, Li Lin, Xu Zhang, Yang Zhang, Wei Zhou, Jinjie Gu 等 9 位作者
The increasing reliance on natural language generation (NLG) models, particularly large language models, has raised concerns about the reliability and accuracy of their outputs. A key challenge is hallucination, where models produce plausible but incorrect information. As a result, hallucination detection has become a critical task. In this work, we introduce a comprehensive hallucination taxonomy with 11 categories across various NLG tasks and propose the HAllucination Detection (HAD) models, which integrate hallucination detection, span-level identification, and correction into a single inference process. Trained on an elaborate synthetic dataset of about 90K samples, our HAD models are versatile and can be applied to various NLG tasks. We also carefully annotate a test set for hallucination detection, called HADTest, which contains 2,248 samples. Evaluations on in-domain and out-of-domain test sets show that our HAD models generally outperform the existing baselines, achieving state-of-the-art results on HaluEval, FactCHD, and FaithBench, confirming their robustness and versatility.
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 6: Industry Track) · 2026 · 正式发表
Cao Te, Mengge Xue, Zhenyu Hu, Yuan Chen, Liqun Liu, Peng Shu, Huan Yu, Jie Jiang
While advertising is a cornerstone of commercial growth, it is constrained by online violation detection systems that reject non-compliant content at a million-scale daily. Advertisers urgently require automated solutions to rectify these advertisements, especially visual ads, as manual fixing is unscalable. Although recent safety-driven methods can achieve compliance, they typically suffer from over-editing, destroying the original commercial intent and perceptual similarity.To address this, we present SSR-A, a framework tailored for the minimalist rectification of non-compliant image ads.Instead of fine-tuning image editing models directly, SSR-A focuses on translating violation policies into targeted editing instructions.We first introduce a Spatial- and Semantic-Aware Instruction Synthesis Pipeline, where MLLMs synthesize candidate instructions—incorporating spatial grounding and semantic guidance—and select the optimal instruction via multi-dimensional evaluation. Furthermore, we align the model using Curriculum Reinforcement Learning, employing GRPO with multi-faceted rewards to progressively navigate the trade-off between compliance and visual preservation. Extensive experiments and online A/B tests show that SSR-A significantly outperforms state-of-the-art baselines in both compliance and preservation of visual and commercial consistency.
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 6: Industry Track) · 2026 · 正式发表
Yuan Chen, Zhenyu Hu, Mengge Xue, Cao Te, Liqun Liu, Peng Shu, Huan Yu, Jie Jiang
Rigorous content moderation is crucial for online advertising but leads to millions of daily rejections. This scale renders manual rectification infeasible, particularly for video advertisements.However, existing safety-driven methods often suffer from aggressive over-editing, which compromises the advertiser’s original semantic intent merely to satisfy compliance.In this work, we target the rectification of textual violations in video ads, covering both speech transcripts and on-screen text. We propose \mathcal{R}^3, a novel framework designed to harmonize compliance with original semantic intent preservation.Our approach integrates three key innovations: (1) an experience-driven data synthesis framework that bootstraps high-quality supervision via group-Relative compliance experience extractor; (2) a curriculum Reinforcement learning strategy with hierarchical rewards designed to enforce compliance while maximizing semantic consistency;and (3) a comprehensive video Rectification framework seamlessly integrating text recognition, rewriting, and re-rendering for industrial deployment. Extensive experiments on industrial datasets and online A/B testing demonstrate that \mathcal{R}^3 significantly outperforms state-of-the-art baselines, achieving an optimal trade-off between violation rectification and intent preservation.
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 6: Industry Track) · 2026 · 正式发表
Hong Ting Tsang, Jiaxin Bai Hong Kong Baptist University, Haoyu Huang, Qiao Xiao, Tianshi Zheng The Hong Kong University of Science and Technology, Baixuan Xu Hong Kong University of Science and Technology, Shujie Liu, Yangqiu Song Hong Kong University of Science and Technology
Building effective knowledge graphs (KGs) for Retrieval-Augmented Generation (RAG) is pivotal for advancing question answering (QA) systems. However, its effectiveness is hindered by a fundamental disconnect: the knowledge graph (KG) construction process is decoupled from its downstream application, yielding suboptimal graph structures. To bridge this gap, we introduce AutoGraph-R1, the first framework to directly optimize KG construction for task performance using Reinforcement Learning (RL). AutoGraph-R1 trains an LLM constructor by framing graph generation as a policy learning problem, where the reward is derived from the graph’s functional utility in a RAG pipeline. We design two novel, task-aware reward functions, one for graphs as knowledge carriers and another as knowledge indices. Across multiple QA benchmarks, AutoGraph-R1 consistently enables graph RAG methods to achieve significant performance gains over using task-agnostic baseline graphs. Our work shows it is possible to close the loop between construction and application, shifting the paradigm from building intrinsically ‘good‘ graphs to building demonstrably ‘useful‘ ones.
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
Yifan Zhang SalesForce.com, Chen Huang National University of Singapore, Yueke Zhang, Jiahao Zhang Vanderbilt University, Toby Jia-Jun Li University of Notre Dame, Collin McMillan University of Notre Dame, Kevin Leach Vanderbilt University, Yu Huang Vanderbilt University
Code Language Models (CodeLLMs) traditionally learn attention based solely on statistical input-output token correlations (“machine attention”). In contrast, human developers rely on intuition, selectively fixating on semantically salient tokens during program comprehension. We present EyeMulator, a model-agnostic technique to align CodeLLM attention with human visual attention without architectural changes. By extracting scan paths from eye-tracking data, we derive token-level attention weights used to augment the loss function during fine-tuning. This induces the model to mimic human focus. Our evaluation across StarCoder, Llama-3.2, and DeepSeek-Coder shows that EyeMulator significantly outperforms baselines, achieving gains of over 30 CodeBLEU points in translation and up to 22 BERTScore points in summarization. Ablation studies confirm that these gains stem directly from replicating human attention dynamics. Artifacts are available at https://zenodo.org/records/17205682.
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
Large Language Models (LLMs) have revolutionized various fields, yet their training efficiency is heavily reliant on effective data curation. While data selection has been widely studied, the strategic data organization for enhanced training remains an underexplored area, particularly since current LLMs are often trained for only one or a few epochs. This paper systematically explores the influence of data organization on LLM training by reusing pre-computed sample-level scores originally generated for data efficiency, thereby incurring minimal additional computational overhead. We identify and formalize four key guidances for optimizing data organization: Boundary Sharpening, Cyclic Scheduling, Curriculum Continuity, and Local Diversity. Guided by them, we introduce two novel data ordering methods termed STR and SAW. Extensive experiments across different model scales and data sizes, encompassing both pre-training and SFT stages, validate the effectiveness of our summarized guidances. They also demonstrate the robustness of our proposed data ordering methods in enhancing the stability and performance of LLM training.
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
Dexuan Xu, Yuan Jiayin, Jianing Wang Peking University, Yanyuan Chen, Hanpin Wang Peking University, Yu Huang Peking University
Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities in interpreting single medical images. However, real-world clinical diagnosis is intrinsically a multi-view process, requiring the synthesis of information across volumetric slices, temporal sequences, and comparative modalities. Existing benchmarks fail to capture this complexity, limiting the assessment of models in realistic clinical workflows. To bridge this gap, we introduce MedMultiBench, the first large-scale benchmark specifically designed for medical multi-image understanding. Comprising 11,392 expert-curated samples, MedMultiBench evaluates MLLMs across four distinct dimensions: Joint Reasoning, Comparative Analysis, Comprehensive Perception, and In-Context Learning. We benchmark 13 state-of-the-art MLLMs, revealing that while current models excel in single-view tasks, they struggle significantly with multi-image contexts. Our experiments identify a performance degradation in open-source models when processing increased visual loads, whereas closed-source models demonstrate better scalability. MedMultiBench provides a robust framework to facilitate the development of MLLMs capable of holistic clinical 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
Tianyu Huang, Yida Zhao, Chuyan Zhou, Kewei Tu ShanghaiTech University
Augmenting Transformers with linguistic structures effectively enhances the syntactic generalization performance of language models. Previous work in this direction focuses on syntactic tree structures of languages, in particular constituency tree structures. We propose Graph-Infused Layers Transformer Language Model (GiLT) which leverages dependency graphs for augmenting Transformer language models. Unlike most previous work, GiLT does not insert extra structural tokens in language modeling; instead, it injects structural information into language modeling by modulating attention weights in the Transformer with features extracted from the dependency graph that is incrementally constructed along with token prediction. In our experiments, GiLT with semantic dependency graphs achieves better syntactic generalization while maintaining competitive perplexity in comparison with Transformer language model baselines. In addition, GiLT can be finetuned from a pretrained language model to achieve improved downstream task performance. Our code is released at https://github.com/cookie-pie-oops/GiLT-LM.
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
Yuxing Lu, Xukai Zhao, J. Ben Tamo, Micky C. Nnamdi, Rui Peng Peking University, Shuang Zeng Georgia Institute of Technology and Peking University, Xingyu Hu, Jinzhuo Wang Peking University 等 9 位作者
Large Language Models (LLMs) have demonstrated remarkable capabilities on general text; however, their proficiency in specialized scientific domains that require deep, interconnected knowledge remains largely uncharacterized. Metabolomics presents unique challenges with its complex biochemical pathways, heterogeneous identifier systems, and fragmented databases. To systematically evaluate LLM capabilities in this domain, we introduce MetaBench, the first benchmark for metabolomics assessment. Curated from authoritative public resources, MetaBench evaluates five capabilities essential for metabolomics research: knowledge, understanding, grounding, reasoning, and research. Our evaluation of 25 open- and closed-source LLMs reveals distinct performance patterns across metabolomics tasks: while models perform well on text generation tasks, cross-database identifier grounding remains challenging even with retrieval augmentation. Model performance also decreases on long-tail metabolites with sparse annotations. With MetaBench, we provide essential infrastructure for developing and evaluating metabolomics AI systems, enabling systematic progress toward reliable computational tools for metabolomics research.
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
Manyi Zhang Huawei Technologies Ltd., Ji-Fu Li Huawei Technologies Ltd., Zhongao Sun Huawei Technologies Ltd., Haoli Bai Huawei Technologies Ltd., Hui-Ling Zhen, Zhenhua Dong, Xianzhi Yu Huawei Technologies Ltd.
Microscaling Floating-Point (MXFP) has emerged as a promising low-precision format for large language models (LLMs). Despite various post-training quantization (PTQ) algorithms being proposed, they mostly focus on integer quantization, while their applicability and behavior under MXFP formats remain largely unexplored. To address this gap, this work conducts a systematic investigation of PTQ under MXFP formats, encompassing over 7 PTQ algorithms, 15 evaluation benchmarks, and 3 LLM families. The key findings include: 1) MXFP8 consistently achieves near-lossless performance, while MXFP4 introduces substantial accuracy degradation and remains challenging; 2) PTQ effectiveness under MXFP depends strongly on format compatibility, with some algorithmic paradigms being consistently more effective than others; 3) PTQ performance exhibits highly consistent trends across model families and modalities, in particular, quantization sensitivity is dominated by the language model rather than the vision encoder in multimodal LLMs; 4) The scaling factor of quantization is a critical error source in MXFP4, and a simple pre-scale optimization strategy can significantly mitigate its impact. Together, these results provide practical guidance on adapting existing PTQ methods to MXFP quantization.
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
Hao Zong, Cong Hu Yuan, Chao Bei, Wentao Chen, Huan Liu Global Tone Communication Technology, Kaiyu Huang Beijing Jiaotong University, Degen Huang Fuyao University of Science and Technology
Current paradigms for empowering Large Language Models (LLMs) with multilingual capabilities rely heavily on massive instruction tuning. We challenge this view, proposing that the barrier is topological alignment, not data quantity. We introduce Hybrid Cross-Alignment (HCA), fusing a frozen NLLB encoder with a Qwen decoder via a closed-loop dual-adapter architecture. HCA utilizes a Source-Side Adapter to precondition encoder features and a Query-Residual Adapter to preserve generative stability, bridged by an adaptive gated cross-modal interface. Our core discovery is Universal Alignment Generalization.” We demonstrate that training HCA on a single language pair (German-English) unlocks state-of-the-art zero-shot transfer to dozens of unseen languages. Crucially, our Oracle” experiments reveal that this single-pair training recovers over 96.7% of the performance achievable by training on all available pairs. This proves that a universal, language-agnostic projection protocol exists. With a total inference footprint of 5.25B parameters, our model significantly outperforms larger baselines, surpassing TowerPlus-9B (+9.0 COMET on low-resource languages) and Aya-101 (13B). Furthermore, performance scales linearly with encoder size; upgrading from 600M to 1.3B yields immediate gains (+3.4 points on Gujarati) with minimal retraining cost.
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
Junpeng Liu, Jiuyi Li Dalian University of Technology, Kaiyu Huang Beijing Jiaotong University, Bo Jin Dalian University of Technology, Degen Huang Fuyao University of Science and Technology, Hui Xiong Hong Kong University of Science and Technology (Guangzhou)
Current large language models (LLMs) often exhibit performance imbalances between dominant languages (e.g., English) and non-dominant ones due to the skewed distribution of pretraining data. A common strategy to address this issue is to enhance cross-lingual alignment, thereby facilitating non-dominant language processing. However, existing methods typically rely on additional training objectives or language-specific parameters, which increase training complexity and cost. In this work, we propose a selective bidirectional language projection framework that enables efficient multilingual alignment and language shift using the intrinsic parameters. Specifically, we first identify the layers most sensitive to language projection between non-dominant and dominant languages through neuron activation analysis. We then perform sequential language projection within the selected layers by mapping non-dominant representations into the dominant language space and reverting them before generation. The bidirectional projection benefits the subsequent instruction tuning in non-dominant languages. Experiments on seven benchmarks demonstrate that our method remarkably enhances the performance of non-dominant languages. Further analyses indicate that our method learns better internal representations and exhibits strong generalization capabilities.
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
Mobile GUI agents powered by LMMs can perceive screens and follow instructions, yet existing benchmarks largely target short, linear workflows and step-level accuracy, offering limited insight into long-horizon planning and decision-making under branching structures. We present DAC-Bench, a decision-aware benchmark with compositional tasks comprising 830 episodes and 11,345 action steps across 35 applications on Android and iOS. Tasks are organized into Sequential, Conjunctive, Conditional, and Hierarchical structures, reflecting real-world multi-step and branching interaction patterns. To complement standard step-level evaluation, we introduce weighted longest common subsequence to capture length-sensitive progress and decision accuracy for branch correctness. Evaluations across 7 diverse agents show substantial performance degradation compared to prior benchmarks, with success rates dropping below 5% on 6–8 step tasks and branch accuracy averaging 38%, highlighting challenges in conditional decision-making. By exposing these failure modes, DAC-Bench provides a challenging and diagnostic benchmark for advancing decision-aware mobile GUI agents. Our code and dataset are available at: https://github.com/YuqingZhangMirror12/DAC-Bench.
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
Zehua Pei, Hui-Ling Zhen, Lancheng Zou The Chinese University of Hong Kong, Xianzhi Yu Huawei Technologies Ltd., Wulong Liu, Sinno Jialin Pan, Mingxuan Yuan, Bei Yu Department of Computer Science and Engineering, The Chinese University of Hong Kong
Scaling large language models (LLMs) improves performance but significantly increases inference costs, with feed-forward networks (FFNs) consuming the majority of computational resources. While Mixture-of-Experts (MoE) architectures can reduce this cost through sparse activation, restructuring existing dense models into MoEs typically requires extensive retraining on hundreds of billions of tokens.We propose an analytical post-training framework that rapidly restructures FFNs into sparse MoE architectures using only a small calibration dataset. The method analyzes neuron activation patterns to partition neurons into always-active shared experts and conditionally activated routed experts, then constructs a router analytically from representative neuron statistics, enabling immediate deployment or optional lightweight fine-tuning. This approach applies both to dense models and recursively to existing MoE models for hierarchical sparsity.Experiments demonstrate up to 1.17\times speedup in compute-bound scenarios with only minutes of processing and 2k-sample fine-tuning, outperforming methods requiring orders of magnitude more resources.
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
Xuchen Li Institute of Automation, Chinese Academy of Sciences and Zhongguancun Academy, Xuzhao Li Beijing Institute of Technology, Jiahui Gao, Renjie Pi, Shiyu Hu Nanyang Technological University, Wentao Zhang
Vision-Language Models (VLMs) excel at many multimodal tasks, yet they frequently struggle with tasks requiring precise understanding and handling of fine-grained visual elements. This is mainly due to information loss during image encoding or insufficient attention to critical regions. Recent work has shown promise by incorporating pixel-level visual information into the reasoning process, enabling VLMs to access high-resolution visual details during their thought process. However, this pixel-level information is often overused, leading to inefficiency and distraction from irrelevant visual details. To address these challenges, we propose the first framework for adaptive pixel reasoning that dynamically determines necessary pixel-level operations based on the input query. Specifically, we first apply operation-aware supervised fine-tuning to establish baseline competence in textual reasoning and visual operations, then design a novel rollout-guided reinforcement learning framework relying on feedback of the model’s own responses, which enables the VLM to determine when pixel operations should be invoked based on query difficulty. Experiments on extensive multimodal reasoning benchmarks show that our model achieves superior performance while significantly reducing unnecessary visual operations. Impressively, our model achieves 73.4% accuracy on HR-Bench 4K while maintaining a tool usage ratio of only 20.1%, improving accuracy and simultaneously reducing tool usage by 66.5% compared to the previous methods.
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
Ruiyao Xu Northwestern University, Mihir Parmar Google, Tiankai Yang University of Southern California, Zhengyu Hu University of Washington, Yue Zhao University of Southern California, Kaize Ding Northwestern University
Learning from preference-based feedback has become an effective approach for aligning LLMs across diverse tasks. However, high-quality human-annotated preference data remains expensive and scarce. Existing methods address this challenge through either self-rewarding, which scales by using purely AI-generated labels but risks unreliability, or active learning, which ensures quality through oracle annotation but cannot fully leverage unlabeled data. In this paper, we present CoAct, a novel framework that synergistically combines self-rewarding and active learning through strategic human-AI collaboration. CoAct leverages self-consistency to identify both reliable self-labeled data and samples requiring oracle verification. Additionally, oracle feedback guides the model to generate new instructions within its solvable capability. Evaluated on three reasoning benchmarks across two model families, CoAct achieves average improvements of +13.25% on GSM8K, +8.19% on MATH, and +13.16% on WebInstruct, consistently outperforming all baselines.
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
Xuan Xiong, Huan Liu Amazon, Li Gu Concordia University, Zhixiang Chi University of Toronto and Huawei Technologies Ltd., Yue Qiu, Yuanhao YU Huawei Technologies Ltd., Yang Wang Concordia University
Chain-of-thought (CoT) reasoning improves large language model performance on complex tasks, but often produces excessively long and inefficient reasoning traces. Existing methods shorten CoTs using length penalties or global entropy reduction, implicitly assuming that low uncertainty is desirable throughout reasoning. We show instead that reasoning efficiency is governed by the trajectory of uncertainty. CoTs with dominant downward entropy trends are substantially shorter. Motivated by this insight, we propose Entropy Trend Reward (ETR), a trajectory-aware objective that encourages progressive uncertainty reduction while allowing limited local exploration. We integrate ETR into Group Relative Policy Optimization (GRPO) and evaluate it across multiple reasoning models and challenging benchmarks. ETR consistently achieves a superior accuracy–efficiency trade-off, improving DeepSeek-R1-Distill-7B by +9.9% accuracy while reducing CoT length by 67% across four benchmarks.
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