ForgetMimic: Motion Unlearning for Reinforcement Learning Humanoid Control
PoEM: Predicting RL Outcomes from Existing Policies
Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures
AV-GRPO: Modality-Anchored Decoupling Diffusion Reinforcement Learning for Joint Audio-Video Generation
Rufus-Air: An Open LLM Post-Training Recipe
IterSynth: Rethinking Deep Search Agents via Role-Decoupled Iterative Synthesis
Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner Agents
When and Where to Trust the Teacher: Unifying On-Policy Distillation and GRPO through Entropy-Calibrated Credit Assignment
LEAP-CBF: A Safety Filter for Uncertain Systems with Least-Effort Adversarial Potentials
Anthropic engineer explains why Claude's writing got worse although the model got smarter
2 min
Kyutai Releases Voice of Reason: A Speech-Native Model that Solves Spoken Math with Reinforcement Learning
4 min
Snorkel AI triples valuation to $3.5B as demand for AI training data booms
2 min
Verifiable Hidden Dynamics Play: Generating Agentic RL Environments from Solved Mechanisms
A Decentralized Partially Observable Team Decision Methodology with Delayed Information Sharing
SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue
Optimal Sequential Annotations for Off-Policy Evaluation
Beyond Repeated Sampling: Learning Search Policies for LLM Reasoning
MAGIC: Mixed-Granularity Agent Graphs via Incremental Construction with Dense-Reward Reinforcement Learning
[AINews] Xiaomi MiMo-V2.6-Pro 1T-A42B: the new top Open Weights model, trained for $3M
5 sources 13 min
PACT: From Credit Assignment to Critic Alignment
