A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Jens Kober, J. Andrew Bagnell, Jan Peters
A comprehensive survey of reinforcement learning for robot behavior generation, highlighting key challenges, successes, and the trade-offs between model-based/model-free and value-function/policy-search methods.
Keming Wu, Baoyi Wang, Kaichen Zhang, et al.
StreamOPD is a post-training recipe combining verifiable streaming-video data, thinking-mode on-policy distillation, and instruct-mode deployment, improving streaming video understanding without inference-time memory or retrieval.
Gokul Swamy, Sanjiban Choudhury, Wen Sun, et al.
This paper investigates why two-stage RL fine-tuning outperforms offline MLE, finding that RL's value lies in searching over a reduced policy space defined by simple verifiers.
Chujie Zheng, Shixuan Liu, Mingze Li, et al.
GSPO defines sequence-level importance ratios and clipping for stable, efficient RL training of LLMs, outperforming GRPO and stabilizing MoE training.
Ebenezer Gelo, Geraud Nangue Tasse, Steven James, et al.
Proposes RCI framework to convert sparse trajectory-level stop-feedback into dense per-step costs for safe offline RL, preserving feasible policy set and optimal Lagrangian.
Vatsal Venkatkrishna, Nico Daheim, Iryna Gurevych
Introduces parameter-space exploration for LLM RL via Perturbed Parameter Policy Optimization (3PO), improving downstream performance over GRPO at similar FLOPs.
Yijiang Li, Bingyang Wang, Yijun Liang, et al.
Proposes U-OPSD, an unsupervised on-policy self-distillation method that uses only a model's own generations and majority voting to improve reasoning without external supervision.
Unknown
This paper presents a Real-to-Sim-to-Real framework that achieves RGB-only sim-to-real policy transfer for visual robot navigation and locomotion.
Unknown
This paper introduces a one-shot imitation learning framework that enables a robot to replicate a demonstrated task from a single visual demonstration using a learned neural network policy.
Unknown
Villa-x enhances latent action modeling in Vision-Language-Action models to improve robot manipulation policy learning and generalization.
Unknown
This paper describes policy iteration for multi-agent reinforcement learning, focusing on an algorithm called Interconnected Learning Automata for Markov Games (MG-ILA).
Unknown
This paper introduces a model-based reinforcement learning approach via meta-policy optimization to improve data efficiency by learning dynamics models that better match the true environment.