Apple ML Proposes SELFCOMPACT: Letting Agents Decide When and How to Compress Their Own Context
Apple ML Research · rss · 2026-10-12
Apple ML Research introduces SELFCOMPACT, a scaffold addressing context pollution in long-running language model agents.
Problem: long agent traces of chain-of-thought and tool calls accumulate stale content that anchors later generations and eventually overflows the context window. Existing scaffolds compact at fixed token thresholds, ignoring trajectory structure and risking discarding partial results mid-derivation or mid-search.
Approach: SELFCOMPACT lets the model itself decide when and how to compact, via two inference-time elements:
- a compaction tool the model invokes to summarize accumulated context into a compact summary;
- accompanying inference-time guidance so compaction happens at structurally appropriate points.
The work targets agent memory/context management and will appear as an Apple ML research paper.
More from coding & agent
- OpenJet claims Qwen3.8-27B coding runs 1.9x faster than llama.cpp on one RTX 3090 — JeremyCMorgan · 2026-10-23
- Building an autonomous SEO agent in Hermes with three skills: programmatic pages, Keywords Everywhere, DataForSEO — EXM7777 · 2026-10-23
- Building a sports betting AI agent with onchain execution and human approval by default: architecture and lessons — Dry-Comfortable-2678 · 2026-10-23
- Capy 0.5.0 lets Claude Pro/Max subscribers reuse their plan quota, adds OIDC cloud access — garrytan · 2026-10-23
- You.com CEO Richard Socher: legacy code is human slop, let AI clean it up — RichardSocher · 2026-10-23
- Let agents design a single-line glance component so you can track them from the sidebar — blixt · 2026-10-23