This paper establishes theoretical bounds on predicting agent behavior from observed actions, based on causal models, highlighting fundamental limits in inferring beliefs in unseen environments.

Alexis Bellot, Jonathan Richens, Tom Everitt

Proposes doubly robust counterfactual policy mean embeddings (CPME) for flexible, nonparametric distribution estimation and hypothesis testing, outperforming existing methods.

Houssam Zenati, Bariscan Bozkurt, Arthur Gretton

Using Stochastic Activity Network model to quantify AV safety at intersections, achieving accident rates below 10^-5 per hour.

Peter Popov, Lorenzo Strigini, Cornelius Buerkle et al.

Fine-tuning OpenAI Whisper with limited MSA data significantly improves recognition of five major Arabic dialects, matching larger models' performance.

Ömer Tarik Özyilmaz, Matt Coler, Matias Valdenegro-Toro

ReasoningFlow analyzes complex reasoning traces' semantic structures using directed acyclic graphs.

Jinu Lee, Sagnik Mukherjee, Dilek Hakkani-Tur et al.

STORYTELLER uses SVO nodes and dynamic modules to improve story coherence, achieving 84.33% human preference rate.

Jiaming Li, Yukun Chen, Ziqiang Liu et al.

Proposes angle concentration as an intrinsic model signal, enabling GAIN-RL to accelerate training over 2.5× with dynamic data scheduling.

Qinsi Wang, Jinghan Ke, Hancheng Ye et al.

Small language models (SLMs) are shown to match or surpass large models in agentic AI tasks, offering lower costs and higher efficiency, with a migration algorithm proposed.

Peter Belcak, Greg Heinrich, Shizhe Diao et al.

SynthRL uses verifiable data synthesis to generate challenging training samples, boosting visual reasoning model performance by +1-2% on benchmarks.

Zijian Wu, Jinjie Ni, Xiangyan Liu et al.

Temporal-RLT framework enhances video understanding efficiency through reward design and data selection, significantly reducing training data.

Hongyu Li, Songhao Han, Yue Liao et al.

ShapeLLM-Omni employs 3D VQVAE and autoregressive transformers to enable unified 3D understanding and generation, supported by the 3D-Alpaca dataset.

Junliang Ye, Zhengyi Wang, Ruowen Zhao et al.

UMA introduces multi-level surface alignment guided by 2D point tracking, achieving 30% reduction in geometric error and 20% detail enhancement at 4K resolution.

Heming Zhu, Guoxing Sun, Christian Theobalt et al.

Introduces Common Corpus, a 2 trillion token open multilingual dataset for ethical LLM pretraining.

Pierre-Carl Langlais, Pavel Chizhov, Catherine Arnett et al.

Introduces Attention-Indexed Model (AIM) to predict Bayes-optimal generalization error using statistical mechanics.

Fabrizio Boncoraglio, Emanuele Troiani, Vittorio Erba et al.

PGPO introduces pseudocode plans and preference optimization, boosting LLM reasoning with 11.6% performance gain.

Zouying Cao, Runze Wang, Yifei Yang et al.

A unified ODE/SDE tutorial on flow matching and diffusion models

Peter Holderrieth, Ezra Erives

HASHIRU enhances resource utilization via CEO-agent hierarchy, achieving 96% on GSM8K.

Kunal Pai, Parth Shah, Harshil Patel

Proposes a dynamic, socially-aware AI alignment framework emphasizing environment-dependent interactions in multi-agent systems.

Florian Carichon, Aditi Khandelwal, Marylou Fauchard et al.

SealQA benchmark evaluates search-augmented LLMs' reasoning under noisy/conflicting info; models perform poorly, accuracy below 50%.

Thinh Pham, Nguyen Nguyen, Pratibha Zunjare et al.

Proposes Conformal Arbitrage, a risk-controlled model routing framework using conformal risk control to balance objectives with finite-sample guarantees, operating solely at API level.

William Overman, Mohsen Bayati