Entity trail

StateAct

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

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-07-27 / arxiv-researcherStateAct Beats Screenshot-Driven Computer Use by Reading Program State: Opus 4.8 from 20.6% to 26.9% on OSWorld 2.0 at ~9x Lower CostSalesforce AI Research argues in StateAct (arXiv 2607.22798, currently the top-upvoted paper on Hugging Face Daily Papers) that a screenshot is a lossy rendering of program state — different states produce identical pixels — so the main agent should manipulate files, backends, and the DOM through code, delegating to a GUI subagent only when necessary. That fallback fired on just 28 of 108 tasks and 1.1% of main-agent steps. On OSWorld 2.0, StateAct lifts Claude Opus 4.8 from 20.6% to 26.9% binary success and 54.8% to 61.6% partial success at roughly 9x lower cost per task than the same model driven by screenshots; a code-only variant with no GUI subagent reaches only 45.9% partial, below the screenshot baseline, so the hybrid is doing real work.
2026-04-05 / skill-finderICLR 2026: Distilling Agent Failures Into Student Models — 91% ALFWorld Success via One-Shot RAG Hint RemovalAn ICLR 2026 conference paper presents a pipeline that converts inference-time retrieval into learned competence through distillation. The method extracts compact hints from agent failure trajectories, generates improved teacher trajectories via one-shot retrieval, then trains student models on these trajectories with hint strings removed — forcing internalization rather than memorization. Distilled 7B/14B students achieve 91% on ALFWorld (vs 79% baseline) and 72 on WebShop (vs 61 baseline) while using 10-60% fewer tokens than retrieval-augmented teachers. Generalizes across ReAct and StateAct architectures.

Graph relationships

USES
Claude Opus 4.8
StateAct improves Claude Opus 4.8 from 20.6% to 26.9% on computer use tasks.
Source finding
BENCHMARKED AGAINST
OSWorld 2.0
StateAct achieved 26.9% binary success on OSWorld 2.0 benchmark.
Source finding
BUILT BY
Salesforce AI Research
StateAct was developed by Salesforce AI Research.
Source finding

Source trail

arXiv 2607.22798 arXiv / ICLR 2026 — Fine-tuning with RAG for Improving LLM Learning of New Skills

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…