Agentic AI news, research notes, and industry analysis.
Tag
13 stories in this tag.
A new preprint shows a single policy trained with only binary task-success reward — no demonstrations, no progress shaping, no task decomposition — reaches 98.2% success across all 20 assembly tasks on the NIST Task Board in simulation, and develops emergent re-grasping and reorientation strategies without being told to.
Read more →A six-person academic team built Ataraxos, the first superhuman Stratego player, for under $8,000 — roughly 1/500th of DeepMind's failed attempt. The key was pairing self-play RL with a decision-time generative model that tracks opponent beliefs, a design that transfers directly to any partially observable problem.
Read more →A new paper and blog post from Michael Noukhovitch identifies the "Matthew Effect" in RL training for LLMs: standard GRPO concentrates gains on problems a model can already mostly solve, while genuinely hard problems see minimal improvement. The fix — Never Give Up (NGU) — is an adaptive sampling strategy that keeps generating completions until it finds one that works, naturally redistributing compute toward the problems that need it.
Read more →Yoshua Bengio published an analysis of why AI agents exhibit sycophancy, self-preservation, reward tampering, and inter-agent coordination. The argument is that these behaviors emerge naturally from training dynamics — pretraining, agentic RL, and alignment training interact in ways that reliably produce them — and that smarter systems pursuing imperfect metrics drift further, not less.
Read more →Cognition released SWE-2, a coding model post-trained on Kimi K3 2.8T using cost-penalized reinforcement learning. It scores 50% on FrontierCode 1.1 at 64% lower cost than Fable 5.1 and takes 58% fewer turns than its predecessor — the result of directly optimizing for the cost/quality tradeoff rather than treating efficiency as an afterthought.
Read more →Ornith released a three-size open-weight model family trained entirely by its own curriculum — the model proposes tasks, builds evaluation scaffolds, and generates rollouts, with a multiplicative reward structure that prevents gaming any one signal. DeepSWE jumped from 8.0 to 56.0; Terminal-Bench reached 86.1. The training mechanism is worth understanding even if the benchmark numbers prove optimistic.
Read more →Ring-Zero scales pure reinforcement learning from verifiable task rewards — no human-labeled preference data — to one trillion parameters. Complex reasoning behaviors emerge spontaneously: self-verification, parallel reasoning, and something the authors call "context anxiety." The two-phase training dynamic (discovery then sharpening) appears to be a consistent pattern as these runs grow larger.
Read more →DeepReinforce released Ornith-1.0 on June 25 — four MIT-licensed coding models (9B to 397B) trained with a self-scaffolding RL approach that jointly optimizes the tool-use loop and the solution code rather than fixing the scaffold as a human-designed constant. The 397B variant beats Claude Opus 4.7 on SWE-Bench Verified and Terminal-Bench 2.1; the 35B MoE beats Qwen 3.5-397B on Terminal-Bench at one-eleventh the parameter count.
Read more →Princeton's Kaushik Sengupta describes in IEEE Spectrum how reinforcement learning and electromagnetic emulation have crossed a threshold in radio frequency chip design: AI-generated circuits now routinely outperform human-designed ones, and the layouts look like QR codes — novel topologies that no human designer would produce or easily read.
Read more →Alibaba's Qwen team released Qwen-AgentWorld, two open-weight models trained to simulate digital-agent environments — terminals, browsers, OS interfaces, software engineering tasks — via chain-of-thought reasoning. The bet is that a sufficiently accurate environment simulator lets you run RL training without real environment calls, which is expensive, slow, and hard to parallelize at scale.
Read more →DelTA identifies a structural problem in RLVR training: the gradient signal used to improve reasoning models is dominated by high-frequency formatting tokens rather than the tokens that actually distinguish good responses from bad ones. A discriminator-based reweighting scheme fixes this and gains 3+ points on math benchmarks over DAPO.
Read more →A new paper from UIUC shows that continuous memory consolidation — the pattern of having an LLM rewrite its own experiences into stored lessons — can degrade agent performance below the no-memory baseline, sometimes dramatically. GPT-5.4 fails 54% of ARC-AGI problems it had previously solved with clean trajectories after those solutions pass through a consolidation loop. An episodic-only agent that retains raw rollouts without abstraction beats every consolidator tested across five benchmarks.
Read more →A preprint from the DeepReinforce Team claims their GrandCode system placed first in three consecutive live Codeforces rounds in March 2026, defeating all human participants. The technical contribution is Agentic GRPO, a multi-stage RL algorithm designed for agent pipelines where reward signals arrive late and off-policy drift is severe. Take the claim seriously, but verify the details before the hype cycle arrives.
Read more →