Signals 4 · free daily AI digest

CliffCompaction: Cost-Efficient Compaction for Long-Horizon Coding Agents

Paper recorded by Signals 4 on 2026-09-22 in cs.AI. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-22 on arXiv · recorded by Signals 4 on 2026-09-23

Category: cs.AI · 人工智能 · first seen 2026-09-23

Abstract

Agents often work on complex problems that require millions of tokens of context, which necessitates compacting across sessions due to limited context windows. We develop CliffCompaction, an autocompaction technique that reduces cost by up to 50% under a bounded context while maintaining or improving performance on Terminal-Bench and achieving new levels of efficiency for test-time scaling and sta

Read on arXiv →

#2 most recent of 360 cs.AI papers we have recorded · ↑ newer: SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dial · ↓ older: SWE-Serve: Benchmarking Agentic Engineering For Production Inference S
Cite this page: CliffCompaction: Cost-Efficient Compaction for Long-Horizon Coding Agents: the #2 most recent of 360 cs.AI papers we have recorded (as of 2026-09-22). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/cliffcompaction-cost-efficient-compaction-for-long-horizon-coding-agents.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.AI papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions