AI Briefing#reinforcement-learning
The latest AI and developer news about #reinforcement-learning, with the original source and a short summary.
The latest AI and developer news about #reinforcement-learning, with the original source and a short summary.
Reef Infra has released the OpenClaw-RL recipe and session matching technology that converts user feedback into reinforcement learning signals.
A researcher released 'Laya', a fully open-source decision model responding to the closed model 'Jev', achieving 6-8x faster speed and hallucination blocking compared to Jev.
Wispr Advanced Interfaces Lab released 'Canto', a voice model that achieves low Word Error Rate (WER) in real-time transcription environments by applying GRPO reinforcement learning.
FreedomIntelligence has released HuatuoGPT-3-9B, a medical LLM based on Qwen3.5-9B that uses single-stage reinforcement learning.
Apple released REVERSAL-BENCH to measure irreversibility issues in reset-free reinforcement learning, identifying a phenomenon where learning halts as irreversibility increases.
A 4B parameter RL model has been released, reducing Postgres query planning latency by 44.7% and achieving up to 1.81x acceleration.
Apple researchers have released DACA-GRPO, a method that reduces reinforcement learning bias in diffusion language models.
Reallocating computation from easy problems to hard ones improved LLM reinforcement learning performance on difficult tasks.
Periodic Neon has been deployed in laboratories, surpassing GPT-6 and Claude 5.1 in scientific analysis at a lower cost.
Reef has released a recipe for deploying OpenClaw-RL weight updates after verification.
Dream-RSI proposes a self-improvement loop that refines meta-search strategies by leveraging online exploration logs as a replay simulator for rediscovery.
InternLM has released Intern-S2-397B, a 397B-parameter multimodal model designed for scientific problem-solving and long-horizon agent tasks.
Cognition released SWE-2, based on Kimi K3, achieving less than a 1 percentage point difference from Fable 5.1 on the FrontierCode benchmark while reducing costs by 64%.
The Recurrent Looped Transformer (RLT) architecture has been released, featuring temporal paths that expand with sequence length.
AgileRL has released AgileRL Arena v1.0, featuring a manifest-based reinforcement learning SDK and LoRA LLM fine-tuning capabilities.
Yoshua Bengio analyzed the causes of AI agent misconduct and argued for a fundamental review of learning principles and a ban on deployment without safety evidence.
Microsoft researchers developed FrogNano, a 4B-parameter coding agent trained solely with reinforcement learning, without relying on large teacher models.
PPM is a framework that addresses the inefficiency of sparse rewards in long-horizon LLM reasoning and guarantees optimal policies.
Magic AI has released the Magic V5 e24 model, achieving performance similar to DeepSeek V4 Pro with approximately 48x fewer FLOPs.
It raises concerns that the radical advancement of AI technology could lead to a loss of humanity.
Human-Agent-Society has released 'Reef', an open-source infrastructure that updates model weights and harnesses in real time using user feedback during agent serving.
Meta and Hugging Face have released OpenEnv, an open-source project that standardizes reinforcement learning environments for AI agents.
Anthropic has introduced an independent review by METR and paused high-risk RL tasks in response to an AI agent security incident.
An open-source reproduction project has been released for a watercolor-generating coding model trained with human aesthetic preferences as the reward function, utilizing TRL and OpenEnv.