DeepSWIP leverages neural materialization and quotient WMC for exact single-world counterfactual inference in neural probabilistic logic programs, achieving a 2.14× speedup.

Saimun Habib, Vaishak Belle, Fengxiang He

MACR framework resolves LLM knowledge conflicts via multi-agent reasoning, significantly improving accuracy.

Huang Peng, Jiuyang Tang, Weixin Zeng et al.

Proposed Multi-Agent Transactive Memory (MATM) enhances heterogeneous agent populations by sharing trajectories, improving success rate by 8% and reducing steps by 0.59 in interactive tasks.

To Eun Kim, Xuhong He, Dishank Jain et al.

Euclid-Omni unifies deduction and algebra through Euclidea, solving 595/601 Geometry3K problems and 16 IMO-AG-30 problems.

Zhaoyu Li, Hangrui Bi, Youyuan Zhang et al.

Proposed DRFLOW benchmark with 7 metrics, evaluating personalized workflow prediction across 100 tasks and 1246 steps, using multi-source evidence integration.

Md Tawkat Islam Khondaker, Raymond Li, Muhammad Abdul-Mageed et al.

Proposes quality-aware self-distillation with correctness gating and confidence scaling, boosting GUI coordinate accuracy by 2.16%.

Jingyuan Huang, Zuming Huang, Yucheng Shi et al.

FllumaOne introduces a multimodal CAD dataset with executable Python programs validated via OpenCASCADE, enabling high-precision geometric and feature-based learning.

Jizong Zhan

SEAGym creates a dynamic evaluation environment for self-evolving LLM agents, comparing ACE, TF-GRPO, and AHE across multiple metrics on Terminal-Bench 2.0 and HLE.

Congjie Zheng, Chuanyi Xue, Bin Liang et al.

OpenClaw-Skill enhances large language models in tool use and multi-step reasoning via Collective Skill Tree Search.

Tianyi Lin, Chuanyu Sun, Jingyi Zhang et al.

Skill-to-LoRA transforms skill text into LoRA adapters, improving pass rate by 5.2 percentage points.

Tianyi Zhang, Zhonghao Qi

AIChilles finds 49 hidden regressions across 30 AI-evolved system programs using differential, constraint-aware search.

Yajie Zhou, Ao Li, Ashwin Silla et al.

WorkflowView leverages LLMs to abstract low-level action sequences into high-level activities with F1=0.90, demonstrating cross-domain generalization.

Gaurav Verma, Scott Counts

VISTA method improves GUI grounding accuracy via multi-view comparison, significantly enhancing Qwen3-VL model performance.

Xinyu Qiu, Yunzhu Zhang, Heng Jia et al.

StreamMemBench evaluates memory's ability to support future tasks from egocentric streams, focusing on evidence retention, initial use, feedback integration, and reuse, with detailed metrics.

Guanming Liu, Yuqi Ren, Hansu Gu et al.

Using large language models (e.g., Claude 4.7) for automated reproducibility assessment in social sciences, matching effect sizes within ±0.05 and supporting conclusions with high accuracy.

Tobias Holtdirk, Pietro Marcolongo, Anna Steinberg Schulten et al.

Nous extracts behavioral parameters from trading data and attempts prompt-based injection to induce cognitive diversity, but results show limited effectiveness.

Haowei Qian

This study compares Diff-in-Means (DiM) and INLP for extracting linear directions controlling model refusal, finding INLP's counterfactual flipping highly effective.

Elisabetta Rocchetti, Alfio Ferrara

This study systematically evaluates multimodal foundation models' perception, reasoning, and tool use in power defect detection, with detailed experimental data.

Quan Quan

RSMeM enhances remote sensing agents' tool usage accuracy by 6% through knowledge-enhanced memory evolution.

Bingxian Wu, Yu Zhang, Zonghao Guo et al.

Using Causal Influence Diagrams (CID), this paper proves the impossibility of training AI to honestly report latent knowledge through feedback.

Korbinian Friedl, Francis Rhys Ward, Paul Yushin Rapoport et al.