Proposes SelectivelyRM, an optimal transport-based framework for robust reward modeling from noisy preferences, outperforming SOTA by 5-10% in error metrics.

Licheng Pan, Haochen Yang, Haoxuan Li et al.

DiBA (Diagonal and Binary Matrix Approximation) compresses neural network weights by decomposing matrices into three diagonal and two binary factors, improving storage efficiency.

Nobutaka Ono

SkillRet is a large-scale long-text skill retrieval benchmark with 17,810 skills, using structured semantic tags to improve LLM retrieval performance.

Hongcheol Cho, Ryangkyung Kang, Youngeun Kim

Introduces knowledge-graph paths as intermediate supervision, improving self-evolving search agents' QA accuracy by 7.3% on multi-hop tasks.

Huyu Wu, Jun Liu, Xiaochi Wei et al.

Using an information-theoretic and geometric framework, the paper reveals that attention mainly reconfigures representations while FFN drives semantic innovation, exposing redundancy in LVLMs.

Gongli Xi, Ye Tian, Mengyu Yang et al.

OTAD combines residual Riemannian metric and Sinkhorn OT to improve audio distance metrics, addressing FAD's invariance and rank-1 dilution issues.

Wonwoo Jeong

SPARK uses knowledge graphs for structured multi-hop reasoning, outperforming unstructured baselines with 93% accuracy on ScienceQA.

Hyobin Park, Taeseop Kim, Dong-Geol Choi

Proposes a linear-constrained diffusion sampling method combining Langevin mixing and information-theoretic bounds, improving image inpainting and super-resolution.

Ahmad Aghapour, Erhan Bayraktar, Asaf Cohen

StageCF introduces interest burn-down diffusion and reports competitive results against generative and diffusion baselines on three datasets.

Yifang Qin, Zhaobin Li, Arisa Watanabe et al.

LineRides employs line-guided reinforcement learning to enable a bicycle robot to perform diverse stunts without demonstrations or explicit timing, using sparse key orientations.

Seungeun Rho, Shamel Fahmi, Jeonghwan Kim et al.

Study on pop and jazz mix ratios for chord generation using Music Transformer, achieving 84.21% pop accuracy and 75.05% jazz improvement.

Jinju Lee

EIG framework enhances multi-agent scientific ideation with learnable edits-and-commits, excelling on AI Idea Bench 2025.

Jiangwen Dong, Bo Li, Wanyu Lin

StoryAlign evaluates reward models for story generation using StoryRMB; StoryReward achieves 66.3% accuracy on StoryRMB.

Haotian Xia, Hao Peng, Yunjia Qi et al.

Introduces Knowledge-Free Correlated Agreement (KFCA) to incentivize federated learning without ground truth.

Leon Witt, Togrul Abbasli, Kentaroh Toyoda et al.

SEI-SHIELD employs self-supervised contrastive learning and iterative sample rescue to robustly identify emitters under label noise, achieving over 90% accuracy on public RF datasets.

Ruixiang Zhang, Zinan Zhou, Yezhuo Zhang et al.

This paper systematically studies latent action supervision in VLA models, comparing four strategies with specific improvements: image-based and action-based formulations.

Yihan Lin, Haoyang Li, Yang Li et al.

Proposes InfoCoordiBridge, a neuro-symbolic framework integrating multi-sensor coordination and verifiable reasoning, boosting autonomous scene understanding reliability.

Shuo Liu, Lei Shi, Haowen Liu et al.

Proposes online natural language feedback to improve language model alignment via ICL and fine-tuning, achieving up to 35% performance recovery with 50x fewer samples.

Christine Ye, Joe Benton

Transforms natural language safety rules into STL specifications, integrating VLM for zero-shot scene understanding in autonomous navigation.

Kristy Sakano, Kalonji Harrington, Mumu Xu

Jordan-RoPE uses complex Jordan blocks for non-semisimple relative positional encoding, outperforming RoPE on WikiText-103.

Yaobo Zhang