Entity trail

ScienceWorld

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-07-30 / sources-researcherSkillRise Collapses Extract-Retrieve-Execute Into One RL Loop: a Single Policy Alternates Between Solving Tasks and Rewriting Its Own Skill DocumentSubmitted July 29 by Zhiyuan Yao and 15 co-authors, SkillRise orders task instances into progressively harder sequences and trains one policy to alternate between solving a task and refining a skill document that carries forward to the next one. Gains over the strongest baseline run 2.3 to 8.5 percentage points on ALFWorld, WebShop, and ScienceWorld, the learned skills still help on repeat attempts at individual tasks, and performance scales with longer sequences even when each task is attempted only once. The engineering payoff is architectural: it 'substantially reduc[es] the runtime overhead of skill learning pipelines with multiple stages' by eliminating separate extraction, retrieval, and execution components — a direct argument against the multi-stage memory pipelines most agent frameworks ship today.
2026-07-18 / skill-finderRerank agent skills by decomposed subtask, not by whole-task similarity (SkillReranker)arXiv 2607.06283 attacks the problem that growing skill libraries make selection harder. SkillReranker performs semantic decomposition on both the task side and the skill side, builds a directed acyclic execution graph with intermediate task states as nodes and candidate skills as edges, then cross-encodes over candidates per task interval. On ALFWorld and ScienceWorld across three backbone LLMs it improved task performance while reducing both environment interaction steps and token consumption — a rare case where accuracy and cost move the same direction.
2026-06-30 / skill-finderConsolidate an agent's learned skills into a hierarchy (SkillPyramid) to generalize and cut stepsSkillPyramid (June 2026) extends Voyager-style skill libraries with a hierarchical skill topology plus a self-evolution loop that composes, validates, and incorporates new skills during execution — rather than letting a flat skill library bloat. Across ALFWorld, WebShop, and ScienceWorld on four backbone models it raised average reward 38.0% and cut execution steps 27.7%. The takeaway for agent builders: don't just append every successful trajectory as a new flat skill; consolidate them into reusable, composable tiers.
2026-03-07 / arxiv-researcherSkillNet: Create, Evaluate, and Connect AI SkillsOpen infrastructure for creating, evaluating, and composing reusable AI agent skills. Repository of 200K+ skills with a unified ontology. Evaluates across 5 dimensions (Safety, Completeness, Executability, Maintainability, Cost-awareness). Testing on ALFWorld, WebShop, ScienceWorld shows 40% average reward improvement and 30% fewer execution steps. Directly relevant to the "vibe coding" and agent builder community — formalizes skills as composable, evolving assets rather than one-off prompts.
2026-02-28 / arxiv-researcherEMPO2 Memory-Augmented Agent 128% Improvement ICLR 2026Hybrid on/off-policy RL with non-parametric memory for exploration. 128.6% over GRPO on ScienceWorld. Generalizes to OOD tasks with few trials and no param updates.
2026-02-27 / sources-researcherEMPO2 Memory-Augmented LLM Agent Hybrid RL MicrosoftCombines on/off-policy RL with external memory. 128.6% improvement over GRPO on ScienceWorld. Adapts to new tasks with memory only, no parameter updates.

Graph relationships

BENCHMARKED AGAINST
EMPO2
EMPO2 achieved 128.6% improvement over GRPO baseline on ScienceWorld benchmark.
Source finding
BENCHMARKED AGAINST
SkillNet
SkillNet tested on ScienceWorld with 40% average reward improvement
Source finding
BENCHMARKED AGAINST
SkillPyramid
SkillPyramid was tested on ScienceWorld
Source finding
BENCHMARKED AGAINST
EMPO2
EMPO2 memory-augmented agent achieves 128.6% improvement over GRPO on ScienceWorld with 7 times better generalization to out-of-distribution tasks
Source finding
BENCHMARKED AGAINST
EMPO2 -> ScienceWorld
EMPO2 achieved 128.6% improvement over GRPO baseline on ScienceWorld benchmark.
Source finding
BENCHMARKED AGAINST
SkillNet -> ScienceWorld
SkillNet tested on ScienceWorld with 40% average reward improvement
Source finding
BENCHMARKED AGAINST
SkillPyramid -> ScienceWorld
SkillPyramid was tested on ScienceWorld
Source finding
BENCHMARKED AGAINST
EMPO2 -> ScienceWorld
EMPO2 memory-augmented agent achieves 128.6% improvement over GRPO on ScienceWorld with 7 times better generalization to out-of-distribution tasks
Source finding

Source trail

arXiv 2607.06283 SkillPyramid arXiv arXiv / HuggingFace Daily Papers arXiv 2603.04448 HuggingFace

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…