Autoresearch All-Topics Update
State
- 当前
agenda的 primary direction 仍是 Agent-Facing Environment Runtime;GUI Grounding Robustness 是 secondary high,RL-based GUI Agent Training 仍被 credit-assignment 拥挤度约束。 - 本轮按 all-topics recent-window pulse 执行:
fetch_and_score.py --date 2026-07-03 --days 3合并 HF、OpenAlex、CVF 后得到 54 篇候选。arXiv API 返回 429,Lexmount key 未设置,但 HF/OpenAlex/CVF 已足够支撑分流。 - 上轮遗留队列有 6 篇;本轮消化最高优先级的
Dockerless,新增AgenticSTS,其余新信号进入 queue/watchlist。
New Signal
1. AFE verifier 不必只能是 hidden oracle
2606-Dockerless 的核心模式是:verifier 先提出少量诊断问题,再让 read-only sub-agents 进入环境找证据,最后由 judge 聚合 verdict。它在 SWE 里用代码文件和 reference patch 工作,但对 AFE 更重要的是抽象:
- verifier 输入不该只是最终状态或一行 success label;
- verifier 可以要求环境提供可审计 evidence object;
- reward / SFT filter 可以来自 evidence-grounded verifier,而不是必须跑昂贵的 hidden test suite。
这提示 AFE-MiniSuite 应把 verifier 设计成三层:agent-safe observations、read-only verifier evidence、hidden oracle label。实验应测“evidence verifier 是否接近 oracle,同时比 prompt-only feedback 更能提升 recovery / stop / avoid-false-completion”。
2. Memory/context 是 runtime contract,不是越长越好
2607-AgenticSTS 对 AFE 的贡献不在 Slay the Spire 2 本身,而在 “每步 prompt 由 typed layers 组成” 这个 contract:
- L1/L2 是 protocol/schema;
- L3 是环境规则;
- L4 是 episodic summaries;
- L5 是 triggered skills;
- raw transcript 不跨决策无限追加。
GUI/CUA runtime 也需要类似分层:visible UI tree / screenshot evidence、recent failed action summary、retrieved app manual、task-progress verifier notes、stop/recovery skills。否则“给 agent 更多历史”既不可控,也难以判断收益来自哪一种证据。
3. 全域候选正在分化为三条后续线索
- Verifier / reward-source:Dockerless、QVal、PolicyGuard、Decodable Is Not Grounded。这里最贴近 AFE 的是 evidence object 和 verifier feedback。
- Long-horizon memory / skill / horizon scaling:AgenticSTS、Agents-A1、Managing Procedural Memory。这条线提醒 AFE 需要 bounded context 与可审计 skill/memory 更新,而不是无限轨迹日志。
- VLA / world-action / embodied transfer:ABot-M0.5、Domain Arithmetic、Does VLA Even Know the Basics、Learning to Move Before Learning to Do。这些对
VLA-Survey有价值,但不应压过当前 primary AFE。
Collected Papers
| Status | Paper | Note / Reason |
|---|---|---|
| digested | Dockerless: Environment-Free Program Verifier for Coding Agents | 2606-Dockerless |
| digested | AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents | 2607-AgenticSTS |
| queued | Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent | horizon scaling / long trajectory infrastructure |
| queued | Does VLA Even Know the Basics? Measuring Commonsense and World Knowledge Retention in VLA Models | VLA knowledge retention |
| queued | Managing Procedural Memory in LLM Agents | memory / skill benchmark |
| queued | QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents | dense reward/verifier evaluation |
| queued-low | ABot-M0.5, Domain Arithmetic, Learning to Move Before Learning to Do | VLA/world-action, lower current priority |
| skipped duplicate | Qwen-AgentWorld, OpenRath, LUMOS, PersonaVLM, GraphVLM, ActiveVLA, CoA-VLA, MoMa-Kitchen, FedVLA | existing notes / recent reports |
Topic Updates
Agent-Facing Environment Runtime
Dockerless + AgenticSTS jointly push AFE from “state API” toward “bounded evidence contract”。一个 AFE observation 不应只是 DOM/UIA dump,而应回答:本步 decision 允许 agent 看到哪些 typed evidence?哪些 evidence 只能 verifier 看?哪些 evidence 会写入 long-horizon memory?哪些 recovery/stop skill 会被触发?
GUI Grounding Robustness
本轮没有新增 grounding-only 精读,但 Decodable Is Not Grounded 仍是下一轮重要候选。它的 blank-image arbiter 与 EvidenceDependence-GUIGrounding 同向,可用于区分“可线性解码”与“真实视觉依赖”。
RL-based GUI Agent Training
TRIAGE 和 QVal 都指向 dense/process reward,但 credit assignment 子方向此前已因拥挤被降级。本轮判断:只保留 QVal 作为 verifier-evaluation 视角,暂不恢复 TRIAGE 深读,除非 Supervisor 明确要重启 credit assignment。
VLA / Embodied / World Model
Does VLA Even Know the Basics? 比 ABot-M0.5 更值得先读,因为它不是又一个 VLA architecture,而是诊断 VLA fine-tuning 后 commonsense/world knowledge retention 的协议。对 VLA survey 来说,这更可能提供跨模型适用的 evaluation insight。
Next Action
下一轮优先级建议:
Decodable Is Not Grounded:补强 GUI/VLM evidence-dependence 诊断。QVal:判断 dense supervision signal 如何被 cheaply evaluated,服务 AFE verifier/reward design。Does VLA Even Know the Basics?:更新 VLA survey 的 evaluation/retention 线索。
不要急着读 Agents-A1,除非需要专门整理 long-horizon trajectory scaling。它可能更像系统工程 scaling report,而不是当前 AFE 机制缺口。