Jiangze Yan

Jiangze Yan

orcid

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2026 in IEEE Access
Verified
Journal Article
Open access
KZKaikai Zhao
Kaikai Zhao
ZLZHAOXIANG LIU
ZHAOXIANG LIU
CJChe Jiang
Che Jiang
PCPing Chen
Ping Chen
XWXin Wang
Xin Wang
BZBowen Zhou
Bowen Zhou

14 July 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
JYJiangze Yan
Jiangze Yan
YSYi Shen
Yi Shen
WZWenjing Zhang
Wenjing Zhang
JHJieyun Huang
Jieyun Huang
PCPing Chen
Ping Chen
SLSHIGUO LIAN
SHIGUO LIAN
Large language model agents are not merely long-prompt applications: they repeatedly prefill growing transcripts that mix instructions, tool schemas, tool observations, persistent memory, scratchpads, and sub-agent outputs, re-submitting the accumulated context on every turn. This creates a full-session optimization problem in which compression ratio alone is insufficient: practical designs must also preserve cacheable prefixes, recover omitted evidence when it is later needed, maintain provenance across privilege boundaries, and avoid delayed-relevance failures where prematurely discarded detail turns out to matter. This survey reframes context compression as lifecycle management for long-horizon agents rather than one-shot prompt shortening. We organize methods by where information is admitted, placed, compacted, recovered, reused, and governed across a session, then map four method families—natural-language, retrieval and offloading, soft-token, and KV-cache compression—together with learned policy-based methods into this lifecycle, rather than treating them as parallel silos. We analyze how production general-purpose agents instantiate these choices in practice, exposing the triggers, protected regions, and recovery sources that determine whether an agent survives a long run, and we synthesize the fast-moving 2025-2026 frontier in which compression becomes a learned action rather than an external scaffold. Finally, we propose an evaluation protocol that jointly reports task success, peak and total tokens, latency, cache behavior, recoverability, and security-relevant provenance, rather than compression ratio in isolation.