❌

Normal view

Received β€” 4 April 2026 ⏭

Predicting LLM Output Length via Entropy-Guided Representations

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Huanyi Xie Β· Yubin Chen Β· Liangyu Wang Β· Lijie Hu Β· Di Wang
3 April 2026 at 12:00
arXiv:2602.11812v2 Announce Type: replace Abstract: The long-tailed distribution of sequence lengths in LLM serving and reinforcement learning (RL) sampling causes significant computational waste due to excessive padding in batched inference. Existing methods rely on auxiliary models for static length prediction, but they incur high overhead, generalize poorly, and fail in stochastic "one-to-many" sampling scenarios. We introduce a lightweight framework that reuses the main model's internal hidden states for efficient length prediction. Our framework features two core components: 1) Entropy-Guided Token Pooling (EGTP), which uses on-the-fly activations and token entropy for highly accurate static prediction with negligible cost, and 2) Progressive Length Prediction (PLP), which dynamically estimates the remaining length at each decoding step to handle stochastic generation. To validate our approach, we build and release ForeLen, a comprehensive benchmark with long-sequence, Chain-of-Thought, and RL data. On ForeLen, EGTP achieves state-of-the-art accuracy, reducing MAE by 29.16\% over the best baseline. Integrating our methods with a length-aware scheduler yields significant end-to-end throughput gains. Our work provides a new technical and evaluation baseline for efficient LLM inference.
Received β€” 2 April 2026 ⏭

Mechanisms Underlying Drought Adaptability in Duolang Sheep Based on Metabolomic and Transcriptomic Analyses

βœ‡Omics In Lung
By: Bingjie Jiang Β· Bin Chen Β· Yaling Yang Β· Yong Chen Β· Wujun Liu
27 March 2026 at 18:00

Biology (Basel). 2026 Mar 12;15(6):461. doi: 10.3390/biology15060461.

ABSTRACT

This study investigates the mechanisms underlying drought adaptability in Duolang sheep, a local breed from two distinct habitats in Xinjiang-an arid southern region and a grassland northern region-aiming to identify key factors driving differential environmental adaptation. Integrated multi-omics analyses were performed, including serum biochemical assays, untargeted metabolomics of perirenal and tail fat tissues, and transcriptomic profiling of lung, liver, and kidney samples. Our results revealed notable differences: (1) serum levels of GSH-Px, IL-2, and IgG were significantly higher in the southern group (p < 0.01); (2) metabolomic analysis identified key differential metabolites, including EPA (involved in unsaturated fatty acid biosynthesis), choline (glycerophospholipid metabolism), L-serine and glutathione (cofactor biosynthesis), and taurine (sulfur metabolism); and (3) transcriptomic analysis revealed significant differential expression of genes such as FGF21 (thermogenesis), CD14 and DUSP2 (MAPK signaling pathway), GOT1 (arginine biosynthesis), and AVPR2 (vasopressin-regulated water reabsorption). Integrative correlation analysis further indicated that glutathione, EPA, GOT1, and CD14 are involved in energy and lipid metabolism, while taurine, AVPR2, and DUSP2 contribute to oxidative stress resistance and immune regulation. These molecular and metabolic adjustments collectively enhance drought adaptability in southern Xinjiang Duolang sheep. In conclusion, adaptation to arid environments requires enhanced antioxidant capacity and immune function, with metabolites such as EPA supporting lipid metabolism and genes such as FGF21 regulating fatty acid oxidation to limit triglyceride accumulation.

PMID:41892221 | PMC:PMC13023672 | DOI:10.3390/biology15060461

Received β€” 26 March 2026 ⏭

From Noisy Labels to Intrinsic Structure: A Geometric-Structural Dual-Guided Framework for Noise-Robust Medical Image Segmentation

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Tao Wang Β· Zhenxuan Zhang Β· Yuanbo Zhou Β· Xinlin Zhang Β· Yuanbin Chen Β· Tao Tan Β· Guang Yang Β· Tong Tong
25 March 2026 at 12:00
arXiv:2509.02419v2 Announce Type: replace-cross Abstract: The effectiveness of convolutional neural networks in medical image segmentation relies on large-scale, high-quality annotations, which are costly and time-consuming to obtain. Even expert-labeled datasets inevitably contain noise arising from subjectivity and coarse delineations, which disrupt feature learning and adversely impact model performance. To address these challenges, this study propose a Geometric-Structural Dual-Guided Network (GSD-Net), which integrates geometric and structural cues to improve robustness against noisy annotations. It incorporates a Geometric Distance-Aware module that dynamically adjusts pixel-level weights using geometric features, thereby strengthening supervision in reliable regions while suppressing noise. A Structure-Guided Label Refinement module further refines labels with structural priors, and a Knowledge Transfer module enriches supervision and improves sensitivity to local details. To comprehensively assess its effectiveness, we evaluated GSD-Net on six publicly available datasets: four containing three types of simulated label noise, and two with multi-expert annotations that reflect real-world subjectivity and labeling inconsistencies. Experimental results demonstrate that GSD-Net achieves state-of-the-art performance under noisy annotations, achieving improvements of 1.58% on Kvasir, 22.76% on Shenzhen, 8.87% on BU-SUC, and 1.77% on BraTS2020 under SR simulated noise. The codes of this study are available at https://github.com/ortonwang/GSD-Net.
Received β€” 11 March 2026 ⏭

Looking Back and Forth: Cross-Image Attention Calibration and Attentive Preference Learning for Multi-Image Hallucination Mitigation

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Xiaochen Yang Β· Hao Fang Β· Jiawei Kong Β· Yaoxin Mao Β· Bin Chen Β· Shu-Tao Xia
10 March 2026 at 12:00
arXiv:2603.07048v1 Announce Type: cross Abstract: Although large vision-language models (LVLMs) have demonstrated remarkable capabilities, they are prone to hallucinations in multi-image tasks. We attribute this issue to limitations in existing attention mechanisms and insufficient cross-image modeling. Inspired by this, we propose a structured hallucination mitigation framework involving Cross-Image Attention calibration and Preference Learning (CAPL). CAPL explicitly enhances inter-image interactions at the architectural level while reinforcing reliance on genuine cross-image evidence during training, thereby improving the model's perception and modeling of cross-image associations. Specifically, we (i) introduce a selectable image token interaction attention mechanism to establish fine-grained cross-image entity alignment and information flow; (ii) design a cross-image modeling-based preference optimization strategy that contrasts reasoning outcomes under full inter-image interaction and those obtained when images are mutually invisible, encouraging the model to ground its predictions in authentic visual evidence and mitigating erroneous inferences driven by textual priors. Experimental results demonstrate that CAPL consistently improves performance across multiple model architectures, achieving stable gains on both multi-image hallucination and general benchmarks. Notably, performance on single-image visual tasks remains stable or slightly improves, indicating strong generalization capability.

SWE-Fuse: Empowering Software Agents via Issue-free Trajectory Learning and Entropy-aware RLVR Training

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Xin-Cheng Wen Β· Binbin Chen Β· Haoxuan Lan Β· Hang Yu Β· Peng Di Β· Cuiyun Gao
10 March 2026 at 12:00
arXiv:2603.07927v1 Announce Type: cross Abstract: Large language models (LLMs) have transformed the software engineering landscape. Recently, numerous LLM-based agents have been developed to address real-world software issue fixing tasks. Despite their state-of-the-art performance, Despite achieving state-of-the-art performance, these agents face a significant challenge: \textbf{Insufficient high-quality issue descriptions.} Real-world datasets often exhibit misalignments between issue descriptions and their corresponding solutions, introducing noise and ambiguity that mislead automated agents and limit their problem-solving effectiveness. We propose \textbf{\textit{SWE-Fuse}}, an issue-description-aware training framework that fuses issue-description-guided and issue-free samples for training SWE agents. It consists of two key modules: (1) An issue-free-driven trajectory learning module for mitigating potentially misleading issue descriptions while enabling the model to learn step-by-step debugging processes; and (2) An entropy-aware RLVR training module, which adaptively adjusts training dynamics through entropy-driven clipping. It applies relaxed clipping under high entropy to encourage exploration, and stricter clipping under low entropy to ensure training stability. We evaluate SWE-Fuse on the widely studied SWE-bench Verified benchmark shows to demonstrate its effectiveness in solving real-world software problems. Specifically, SWE-Fuse outperforms the best 8B and 32B baselines by 43.0\% and 60.2\% in solve rate, respectively. Furthermore, integrating SWE-Fuse with test-time scaling (TTS) enables further performance improvements, achieving solve rates of 49.8\% and 65.2\% under TTS@8 for the 8B and 32B models, respectively.
Received β€” 25 February 2026 ⏭

MAS-FIRE: Fault Injection and Reliability Evaluation for LLM-Based Multi-Agent Systems

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Jin Jia Β· Zhiling Deng Β· Zhuangbin Chen Β· Yingqi Wang Β· Zibin Zheng
24 February 2026 at 13:00
arXiv:2602.19843v1 Announce Type: cross Abstract: As LLM-based Multi-Agent Systems (MAS) are increasingly deployed for complex tasks, ensuring their reliability has become a pressing challenge. Since MAS coordinate through unstructured natural language rather than rigid protocols, they are prone to semantic failures (e.g., hallucinations, misinterpreted instructions, and reasoning drift) that propagate silently without raising runtime exceptions. Prevailing evaluation approaches, which measure only end-to-end task success, offer limited insight into how these failures arise or how effectively agents recover from them. To bridge this gap, we propose MAS-FIRE, a systematic framework for fault injection and reliability evaluation of MAS. We define a taxonomy of 15 fault types covering intra-agent cognitive errors and inter-agent coordination failures, and inject them via three non-invasive mechanisms: prompt modification, response rewriting, and message routing manipulation. Applying MAS-FIRE to three representative MAS architectures, we uncover a rich set of fault-tolerant behaviors that we organize into four tiers: mechanism, rule, prompt, and reasoning. This tiered view enables fine-grained diagnosis of where and why systems succeed or fail. Our findings reveal that stronger foundation models do not uniformly improve robustness. We further show that architectural topology plays an equally decisive role, with iterative, closed-loop designs neutralizing over 40% of faults that cause catastrophic collapse in linear workflows. MAS-FIRE provides the process-level observability and actionable guidance needed to systematically improve multi-agent systems.
Received β€” 17 February 2026 ⏭

Detecting Brick Kiln Infrastructure at Scale: Graph, Foundation, and Remote Sensing Models for Satellite Imagery Data

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Usman Nazir Β· Xidong Chen Β· Hafiz Muhammad Abubakar Β· Hadia Abu Bakar Β· Raahim Arbaz Β· Fezan Rasool Β· Bin Chen Β· Sara Khalid
17 February 2026 at 13:00
arXiv:2602.13350v1 Announce Type: cross Abstract: Brick kilns are a major source of air pollution and forced labor in South Asia, yet large-scale monitoring remains limited by sparse and outdated ground data. We study brick kiln detection at scale using high-resolution satellite imagery and curate a multi city zoom-20 (0.149 meters per pixel) resolution dataset comprising over 1.3 million image tiles across five regions in South and Central Asia. We propose ClimateGraph, a region-adaptive graph-based model that captures spatial and directional structure in kiln layouts, and evaluate it against established graph learning baselines. In parallel, we assess a remote sensing based detection pipeline and benchmark it against recent foundation models for satellite imagery. Our results highlight complementary strengths across graph, foundation, and remote sensing approaches, providing practical guidance for scalable brick kiln monitoring from satellite imagery.

GREPO: A Benchmark for Graph Neural Networks on Repository-Level Bug Localization

βœ‡cs.AI, q-bio.NC updates on arXiv.org
By: Juntong Wang Β· Libin Chen Β· Xiyuan Wang Β· Shijia Kang Β· Haotong Yang Β· Da Zheng Β· Muhan Zhang
17 February 2026 at 13:00
arXiv:2602.13921v1 Announce Type: cross Abstract: Repository-level bug localization-the task of identifying where code must be modified to fix a bug-is a critical software engineering challenge. Standard Large Language Modles (LLMs) are often unsuitable for this task due to context window limitations that prevent them from processing entire code repositories. As a result, various retrieval methods are commonly used, including keyword matching, text similarity, and simple graph-based heuristics such as Breadth-First Search. Graph Neural Networks (GNNs) offer a promising alternative due to their ability to model complex, repository-wide dependencies; however, their application has been hindered by the lack of a dedicated benchmark. To address this gap, we introduce GREPO, the first GNN benchmark for repository-scale bug localization tasks. GREPO comprises 86 Python repositories and 47294 bug-fixing tasks, providing graph-based data structures ready for direct GNN processing. Our evaluation of various GNN architectures shows outstanding performance compared to established information retrieval baselines. This work highlights the potential of GNNs for bug localization and established GREPO as a foundation resource for future research, The code is available at https://github.com/qingpingmo/GREPO.
There are no more articles
❌