Regression accumulation affects 40-73% of 8-turn LLM coding tasks on extended HumanEval+/MBPP+ benchmarks, with verification gates improving final-turn pass rates on prior tests.
Rossi, Viet Dac Lai, David Seunghyun Yoon, Dilek Hakkani-Tür, and Trung Bui
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5representative citing papers
ContextEcho benchmark shows persona drift occurs across 23 frontier models in long agentic-coding sessions, is not reliably reset by compaction, and can be restored by single-shot anchors with mode-dependent effects.
LLMs exhibit an accumulated message effect where conversation history polarity biases subsequent judgments, stronger for high-entropy items, independent of context length, and with a negativity bias.
Attention to goal tokens declines in multi-turn LLM interactions while residual representations often retain decodable goal information, and the gap between these predicts whether goal-conditioned behavior survives.
Context drift between agents causes hallucinations in multi-agent LLMs; the Shared State Verification Protocol reduces them more effectively than full-broadcast synchronization with 58% fewer API calls.
citing papers explorer
-
Regression Accumulation in Multi-Turn LLM Programming Conversations
Regression accumulation affects 40-73% of 8-turn LLM coding tasks on extended HumanEval+/MBPP+ benchmarks, with verification gates improving final-turn pass rates on prior tests.