Papers
arxiv:2605.24279

ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions

Published on May 22
Authors:
,
,
,

Abstract

Persona stability in language models deteriorates during extended coding sessions, revealing drift that existing benchmarks fail to capture, thus necessitating ContextEcho for comprehensive deployment-scale evaluation.

A frontier language model's acknowledged "helpful programming assistant" persona does not survive long agentic-coding sessions in the deployment regime that production products actually run. After hours of tool-using debugging, a model that initially hedges preferences ("I don't have preferences") may begin asserting them ("Python - the feedback loop is instant..."), revealing user-visible drift that deployer evaluations may miss. Existing persona-stability studies focus on short dialogues and report little shift, leaving real-world code-generation regimes - thousands of tool-using turns, compaction, and hours-long sessions - largely uncharacterized. We introduce ContextEcho, a benchmark and reusable harness for measuring persona drift at deployment scale. It combines a 25-probe identity suite, a snapshot-then-probe protocol that forks conversation state without perturbing the main session, complementary judged and judge-free measurement surfaces, and three anonymized Claude Code sessions spanning 3,746-9,716 turns. Across 23 frontier models, ContextEcho shows that persona drift is general across organizations rather than family-specific, that in-session compaction does not reliably reset it, and that a single-shot anchor restores the trained register across measured targets. It also reveals mode-dependent downstream effects: while drift can facilitate tool-using continuation, in tool-free chat it breaks formatting contracts and inflates output length. Overall, ContextEcho provides researchers and deployers an open-source framework to audit whether the persona a model ships with is the persona users encounter at session end, across chat-completions API targets and without retraining.

Community

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2605.24279
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2605.24279 in a model README.md to link it from this page.

Datasets citing this paper 1

Spaces citing this paper 1

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.