EvoGroup: A Self-Evolving Harness for Group Agents in Shared Contexts

Xinyi Xu; Xiaoyang Li; Yucheng Feng; Wenqing Chen; Youliang Yuan; Wei Shi; Rongxiang Weng; Zibin Zheng

cs.CLNov 2, 2026

EvoGroup: A Self-Evolving Harness for Group Agents in Shared Contexts

Authors: Xinyi Xu, Xiaoyang Li, Yucheng Feng, Wenqing Chen, Youliang Yuan, Wei Shi, Rongxiang Weng, Zibin Zheng

Organizations: Sun Yat-sen University · Longcat Team, Meituan Inc.

Abstract

With the rise of Group Agent such as Claude Tag, agents increasingly serve as digital workers for long-term team collaboration. These agents operate over shared contexts that continuously accumulate through multi-user interactions. However, existing agent memory systems are mainly designed for individual users and struggle to capture group-level knowledge in shared contexts, including group relationships and collaboratively formed knowledge. Manual memory strategies further limit adaptation to diverse and evolving shared contexts. To this end, we propose EvoGroup (EvoG), a self-evolving harness for group agents that starts from a minimal six-tool cold start and adapts through interaction with shared context. EvoG uses confidence and task outcomes to select informative trajectories, while also eliciting self-reflections on uncertainty in low-confidence cases. The analysis agent then examines selected trajectories through adaptive progressive disclosure, followed by bucketed analysis to synthesize high-level findings that guide the evolve agent in revising the Harness. Empirically, pass@1 on held-out split of EverMemBench improves from 71.07% to 89.58%, surpassing both self-evolving memory and human-designed harnesses, including Claude Code (85.83%). To evaluate generalization, we freeze the evolved harness and evaluate it on the unseen GroupMemBench, where, with GPT-5.5, it demonstrates effective generalization and outperforms both memory baselines and human-designed harnesses.

Figures & tables

Figure 1: Performance of EvoGroup with DeepSeek-V4-Flash on EverMemBench . (a) Evolution over 5 iterations on the held-in split, improving pass@1 by $18.2$ pp from $H_{0}$ to $H_{*}$ . (b) Comparison on the held-out split, where $H_{*}$ outperforms all the baselines including Codex and Claude Code.

Figure 2: Overview of the EvoGroup loop , which operates in four phases: I) Interaction, which Group Agent collaborates with shared contexts; II) Trajectory Selection , which thresholds trajectories by confidence; III) Analysis, which performs per-trajectory and bucketed analysis; and IV) Revision, which updates harness through four interfaces.

Panel of Table 1 — Table 1: Performance on EverMemBench. All category scores are aligned to the 1,680-question held-out counts (Single=149, Multi=174, Temp=210, Const=282, Proact=299, Update=188, Style=123, Skill=118, Role=137). Baseline entries are count-aligned projections from the reported full-benchmark category accuracies, not rerun results on the held-out question IDs. Average is recomputed from the aligned integer correct counts. Bold marks the best aligned result within each answer-model block, excluding FullContext.

Figure 3: Component ablation on the held-out EverMemBench split using DeepSeek-V4-Flash. Each variant uses one evolved component, while

H_{*}

uses all five.

\Delta

indicates the gains over

H_{0}

Figure 4: Process ablation on the held-in EverMemBench split using DeepSeek-V4-Flash. Pass@1 under three ablation settings: w/o Trajectory Selection, w/o APD Analysis, and w/o Bucketed Analysis.

Transition	Fail $\rightarrow$ Pass	Pass $\rightarrow$ Fail	Always $\rightarrow$ Pass	Always $\rightarrow$ Fail	Retention $\uparrow$	Regression $\downarrow$
$H_{0}\rightarrow H_{1}$	124	34	469	93	93.24%	6.76%
$H_{1}\rightarrow H_{2}$	73	32	561	54	94.60%	5.40%
$H_{2}\rightarrow H_{3}$	38	53	581	48	91.64%	8.36%
$H_{3}\rightarrow H_{4}$	45	49	570	56	92.08%	7.92%
$H_{4}\rightarrow H_{5}$	53	44	571	52	92.85%	7.15%

Table 2: Capability-transition analysis of self-evolution on the held-in EverMemBench under (a) DeepSeek-V4-Flash and (b) GPT-5.5. Fail

\rightarrow

Pass and Pass

\rightarrow

Fail denote questions from incorrect to correct and from correct to incorrect, respectively. Always

\rightarrow

Pass and Always

\rightarrow

Fail denote questions that remain consistently correct or incorrect across consecutive harnesses. Retention is the fraction of previously correct questions that remain correct, while Regression is the fraction that become incorrect in the next harness.

Iteration	Time (min)		Tokens (M)
Iteration	APD	Full	APD	Full
Iteration 1	64.6	110.8	54.58	82.09
Iteration 2	26.1	43.0	33.34	66.55
Iteration 3	21.5	37.9	22.08	34.18
Iteration 4	17.7	30.1	28.46	56.12
Iteration 5	24.9	47.5	23.88	38.64

Table 3: Effectiveness of Adaptive Progressive Disclosure (APD) for trajectory analysis. We compare APD Analysis with Full Trace Analysis across self-evolution iterations in terms of analysis time and token consumption. APD consistently reduces both analysis time and token consumption compared with Full Trace Analysis, demonstrating its effectiveness in improving trajectory-analysis efficiency. Time is reported in minutes (min), and token usage in millions of tokens (M).

Figure 5: Agreement between self-reflection and trajectory-based diagnosis . Localization Agreement indicates whether self-reflection correctly locates the problematic region, Cause-Attribution Agreement whether it identifies the underlying cause, and Diagnostic-Evidence Consistency whether its diagnosis is supported by trajectory evidence.

Figure 6: Revision-interface edits across evolution rounds for DeepSeek-V4-Flash and GPT-5.5. The figure illustrates which revision interfaces EvoGroup relies on most during self-evolution.

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Panel of Table 4 — Table 4: Performance on GroupMemBench. All metrics are computed using the official scripts and reported as Pass@1 (%). Average is the reported aggregate score. Bold marks the best result within each answer-model block, excluding FullContext. “-” indicates that the input exceeds the model’s context limit.

Iter.	Correct.	Adoption			Repair
Iter.	Correct.	Adopted	Denied	Unused	Repair
Iter. 1	87.50	62.50	0.00	37.50	59.40
Iter. 2	88.90	66.70	0.00	33.30	57.10
Iter. 3	100.00	57.10	14.30	28.60	41.40
Iter. 4	100.00	66.70	0.00	33.30	54.50
Iter. 5	71.40	57.10	14.30	28.60	57.60

Appendix

Table 5: Effectiveness of bucketed analysis in producing useful evolution signals. Correct. measures the consistency between the analyzed findings and the actual task issues; Adopted indicates that the Evolve Agent accepts a finding and uses it to update the harness; Denied indicates that the Evolve Agent considers the finding incorrect and explicitly rejects it; and Unused indicates that the finding is not selected for a harness update. Repair measures the proportion of affected tasks successfully repaired by adopted findings. All the metrics are reported on a 0–100 scale.

Figure 7: Distribution of trajectory types across self-evolution for DeepSeek-V4-Flash and GPT-5.5. CC denotes confident-and-correct trajectories, representing routine successful behavior. CW denotes confident-but-wrong trajectories, revealing potentially undetected reasoning failures or systematic answer bias. UC denotes unconfident-but-correct trajectories, capturing fragile successes that the current harness may not sufficiently support. UW denotes unconfident-and-wrong trajectories, capturing failures that the agent can partially recognize.

Panel of Table 6 — Table 6: Efficiency on EverMemBench . Tok. (K) denotes the average number of tokens consumed per question (in thousands), Calls denotes the average number of tool calls per question, QA denotes the average time required to complete each question, and Cold denotes the one-time preparation cost before answering, i.e., the time required for the initial Harness execution. QA and Cold are measured in seconds.

Agent Harness Self-Evolution

See all 46 papers

MemPro: Agentic Memory Systems as Evolvable Programs
May 30, 2026Qingshan Liu, Guoqing Wang, Wen Wu +5Memory-Augmented Agents Agent Memory Management
East China Normal University · Xiaohongshu Inc.
EvolveNet: Collaborative Harness Evolution for Agent Self-Improvement
Aug 5, 2026Jun Nie, Yonggang Zhang, Qianshu Cai +3Agent Harness Optimization Multi-Agent Collaboration
Hong Kong Baptist University · University of Science and Technology of China · The Hong Kong University of Science and Technology
EvoMemBench: Benchmarking Agent Memory from a Self-Evolving Perspective
May 18, 2026Yuyao Wang, Zhongjian Zhang, Mo Chi +7LLM Agent Memory LLM Agent Evaluation
Hong Kong University of Science and Technology (Guangzhou) · Beijing University of Posts and Telecommunications · Beijing Institute of Technology +1

May 30, 2026J/K move · Enter open · S save

MemPro: Agentic Memory Systems as Evolvable Programs

Qingshan Liu, Guoqing Wang, Wen Wu, Jingqi Huang+4

East China Normal University · Xiaohongshu Inc.

Memory-Augmented Agents Agent Memory Management Self-Evolving Agents

Open paper AskPDF arXiv

Long-horizon autonomous agents require memory systems to retain historical information, track evolving states, and reuse relevant knowledge beyond finite context windows. Existing agentic memory systems typically follow a memory construction-retrieval (MCR) pipeline, but often adapt mainly the memory bank while keeping the surrounding pipeline fixed after deployment. This fixed-pipeline design struggles to handle heterogeneous task-specific failure modes and can become misaligned with memory banks that evolve in scale and structure over time. To address these limitations, we propose MemPro, a system-level evolution framework that treats the entire MCR pipeline as an evolvable program rather than adapting only the memory bank or prompt text. MemPro maintains a version tree of runnable memory-system implementations, where an Evolving Agent iteratively selects promising versions, diagnoses recurring failures, and creates improved child versions through failure-mode-guided edit-debug refinement. Experiments on LongMemEval, LoCoMo, HotpotQA, and NarrativeQA show that MemPro consistently outperforms strong static and prompt-level evolving baselines within a few iterations, continues to improve with evolution, and achieves a favorable performance-cost trade-off. Code is available at https://github.com/wanghai673/MemPro.

Abstract

Agent Harness Self-Evolution

Explore similar work

MemPro: Agentic Memory Systems as Evolvable Programs

EvolveNet: Collaborative Harness Evolution for Agent Self-Improvement

EvoMemBench: Benchmarking Agent Memory from a Self-Evolving Perspective