Proposes a geometric contrastive pretraining framework for GNNs, improving TSP solutions by 6.57% on 1000-node instances.

David Aguado, Daniel Fuertes, Carlos R. del-Blanco et al.

TARA employs type-aware repair allocation, boosting semantic accuracy by +5.6 points on DSG, +2.6 on TIFA, with no retraining needed.

Haoyue Liu, Xiaoyu Ma, Ye Chen et al.

LaT uses a pretrained large language model as an external trainer, enhancing multi-task VRP solutions by stage-wise validation analysis, with minimal overhead.

Yang Wang, Ya-Hui Jia, Wei-Neng Chen et al.

CoDID achieves disentangled learning under hidden correlations via iterative mode discovery and meta-optimization, improving accuracy by 7.8%.

Rong Hu, Ling Chen

This study systematically evaluates trajectory data filtering for LoRA fine-tuning of code agents, revealing data quantity outweighs quality at small scales.

Yunze Han

Proposes diversity-oriented fine-tuning (SFT and DPO) to improve uncertainty-based hallucination detection, boosting AUROC by ~4.5%.

Qiuyuan Li, Hongliang Dai, Piji Li

TTCov uses test-side info to select training data, improving autonomous driving performance by 10% over baselines.

Nadine Chang, Maying Shen, Shizhe Diao et al.

Plover introduces a plan-centric vision-based GUI automation system with externalized, inspectable, and editable task plans, enhancing transparency and control.

Madhumitha Venkatesan, Shicheng Wen, Jiajing Guo et al.

Proposes MCPEvol-Bench, using 11 mutation operators to simulate MCP server evolution, evaluating 12 LLMs' adaptability in dynamic tool environments.

Huanxi Liu, Kun Hu, Jiaqi Liao et al.

MathCoPilot integrates human-in-the-loop proof blueprinting, multi-skill orchestration, and automated knowledge base construction, achieving 20/63 success on FormalMATH with natural language prior to formalization.

Junjie Zhang, Jiayu Liu, Wenbin Liu et al.

Depth-recurrent Transformer achieves per-token fixed-point convergence, reducing average depth by 38%.

Joe Logan

Proposes pairwise validator as reward substitute in self-evolving agents, matching or surpassing full reward performance with reduced labeling cost.

Minghao Liu, Yu Wang, Jiayun Wang et al.

Proposes ASOC framework integrating engineering principles for trustworthy autonomous AI agents.

Amin Beheshti, Rong N. Chang, Boualem Benatallah et al.

Proposes environment-augmented self-evolving medical agents, enhancing autonomy and reliability in clinical settings.

Chunzheng Zhu, Lei Tian, Bohan Tan et al.

Introduces E-P-R framework to diagnose how AI agents consume conflicting memory, revealing the 'compliance trap' phenomenon.

Yixiong Chen, Xinyi Bai, Alan Yuille

Agentic-DPO leverages expert trajectories with preference contrast for offline policy optimization, doubling accuracy from 21.7% to 41.4%.

Yixiong Chen, Alan Yuille

Proposes Who&When Pro, using error injection to generate 12,326 high-quality failure traces across multimodal tasks for automated failure attribution.

Jiale Liu, Huajun Xi, Shaokun Zhang et al.

Introduces Long-Horizon-Terminal-Bench with 46 long-horizon tasks, emphasizing dense rewards and partial credit, to evaluate models' sustained performance.

Zongxia Li, Zhongzhi Li, Yucheng Shi et al.

Introduces self-patching to diagnose and recover 58-75% of reasoning failure caused by knowledge-circuit misalignment in fine-tuned LLMs.

Lu Dai, Ziyang Rao, Yili Wang et al.

MobiDiff uses multi-channel discrete diffusion to efficiently generate privacy-preserving human mobility data.

Rongchao Xu, Lin Jiang, Dahai Yu et al.