Author

Egor Pakhomov

3 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#artificial intelligence Preprint Oct 2026

Does an Agent's History Tell You When Compaction Will Hurt? A Modest, Bounded Effect on the TRACE Paired-Replay Corpus

Many long-horizon agents compact their context on a global rule, usually a token budget, blind to what the agent was doing. We ask whether the agent's recent behaviour predicts when a compaction will hurt. TRACE's public corpus of 590 harness-triggered AppWorld compaction boundaries replays each boundary from a re-exec...

Egor Pakhomov, E. Nijkamp · 0 citations

#artificial intelligence Review Sep 2026

An Architecture for Long-Horizon Agents: Levels, Ticks and Cascaded Intelligence

It is argued that a long-horizon agent must run continually without forgetting before it can learn continually, and that continual learning for these agents needs a substrate outliving every context and process.

E. Nijkamp, Anurag Koul, Egor Pakhomov et al. · 0 citations

#artificial intelligence Preprint Sep 2026

Parsing the Stream: A Live Trace Model for Long-Horizon Agents and Their Observers

A live trace model, an append-only event ledger folded incrementally into typed run state and compiled into per-consumer views, is presented and evaluated for both consumers against deterministic ground truth.

Egor Pakhomov, Erik Nijkamp · 0 citations