LOOPSBENCH introduces a long-horizon benchmark for loop engineering, evaluating coding agents on dependency DAGs with flow-aware runtime, revealing persistent challenges.

Han Li, Zhemin Fang, Rili Feng et al.

AutoCause automates multi-method causal discovery in environmental time series, improving reproducibility and auditability.

Marco Ruiz, Miguel Arana-Catania, David R. Ardila et al.

Introduced CoCo method for expert model interpretation via contribution contrast, enhancing interpretability and accuracy.

Yifan Wang, Jinyi Mu, Mayank Jobanputra et al.

Proposes STAIR, a hierarchical trajectory abstraction framework, achieving 81.2% Pass@1 on SWE-bench with effective knowledge reuse.

Yisen Xu, Jiayuan Zhou, Ruiqi Pan et al.

Decode-Branch Transformer decouples prompt processing from decoding, reducing inference costs and improving performance.

Liming Liu, Mingze Wang, Tuo Zhao

Introduced COMPINT suite to address session constraint loss in context compaction, achieving 90% retention.

Zhiqi Wang, Yichi Zhang, Dongwon Lee et al.

Physics-aware PnP-FBS aligns MMSE noise with the forward model and proves pointwise and Wasserstein recovery bounds.

Tobias Wolf, Jalal Fadili, Jin Guo et al.

OnlineCache enhances diffusion inference efficiency with dynamic caching and error correction, achieving nearly 3x speedup on FLUX.1-dev.

Zhikang Xie, Xichen Ye, Yifan Wu et al.

BWM integrates action-aligned data construction with diffusion-based autoregressive prediction, enhancing robot simulation fidelity and policy evaluation.

BWM Team

Data Turnstile framework generates high-quality function-calling data, boosting Qwen3-0.6B accuracy to 75.9%.

Goutham Ramakrishnan, Megha Sharma

Q-CARE introduces query coverage and claim verifiability for reference-free, fine-grained RAG evaluation, outperforming existing metrics.

Jeonghwan Choi, Taewon Yun, Minjeong Ban et al.

WILC leverages complementarity in LLM collaboration, matching GPT-5.2 performance at 7× lower cost.

Yanbin Fang, Xuan Wei, Wei Chen

DASH-OPD introduces a discrepancy-aware hysteretic switching mechanism for on-policy distillation, improving multi-turn agent training efficiency and performance.

Yuchen Xia, Qianguo Sun, Chao Song et al.

MOSAIC employs adversarial co-evolution with structure-based feature grids, significantly improving heuristic performance for combinatorial optimization problems.

Oguzhan Gungordu, Siheng Xiong, Faramarz Fekri

MESS combines binary hashing, multi-graph HNSW, and differential privacy to enable fast, private semantic search with low latency.

Haoyu Cui, Zengpeng Li, Tien Tuan Anh Dinh et al.

AOCT-MSQR enables training-free video attribution by retrieval, reaching 84.6% Rank-1/78.3% mAP on GenVidBench at 100-shot.

Renxi Cheng, Chaolei Han, Jie Gui et al.

CFTD uses MACE features to evaluate material generation models for quality and novelty.

Paul Hagemann, Katharina Ueltzen, Simon Müller et al.

Study on inference-time scaling in local CUAs shows diminishing returns with increased computation.

Woongkyu Lee, Jungwook Choi

Semantic Similarity Rating (SSR) significantly reduces mode collapse in silicon sampling, improving distribution fidelity with minimal parameters.

Oscar Heath, Rohan Alexander

AIx4Soccer integrates football club management and athlete development using PDI framework and Tak Tik marketplace.

Frederico Falconi Costa, Salvador Cesar Costa, Fabricio F. Costa