When Should Agents Remember? Falsification-Gated Self-Evolution for LLM Agents
Organizations: Nanyang Technological University · Georgia Institute of Technology · University of Washington · Washington University in St. Louis ·
Abstract
LLM agents increasingly operate in long-horizon, verifier-rich environments, but current self-improvement mechanisms often write episode-derived lessons directly into memory, making local success an unreliable signal for reusable and safe knowledge. This paper addresses the core question of when an experience-derived behavioral update should be admitted into persistent agent state. We propose Falsification-Gated Self-Evolution (FGSE), a verifier-grounded framework that represents each candidate lesson as a structured hypothesis with an explicit precondition, behavior change, expected effect, and verifier. FGSE applies the hypothesis only in a temporary state, tests it on target-transfer, falsification, and archived-regression probes, and then commits, refines, or rejects it according to measured gain and risk. Across web, app, tool-use, and long-term memory benchmarks, FGSE achieves strong task performance, including 39.8 WebArena success rate, 78.8 AppWorld average completion, and 0.724/0.487 Pass on -Retail/-Airline, while reducing falsification failures to 5.0%, regression damage to 1.3%, and harmful committed updates to 3.6%. These results suggest that testing memory updates as hypotheses supports more reliable long-term adaptation.
Figures & tables
| Method | Backbone / Setting | WebArena SR | AppWorld Normal | AppWorld Challenge | AppWorld Avg. | -Retail Pass 1 | -Airline Pass 1 |
| Web / browser-control agents: methods primarily designed or evaluated for browser navigation and web interaction. | |||||||
| GPT-4 Agent ( Zhou et al., 2024 ) | GPT-4 | 14.4 | 43.2 | 25.4 | 34.3 | 0.553 | 0.378 |
| AutoEval ( Pan et al., 2024 ) | GPT-4 | 20.2 | 45.8 | 28.7 | 37.3 | 0.579 | 0.391 |
| BrowserGym ( Le Sellier de Chezelles et al., 2025 ) | GPT-4o | 24.3 | 46.5 | 29.1 | 37.8 | 0.585 | 0.402 |
| CER online ( Liu et al., 2025 ) | GPT-4o | 33.2 | 51.9 | 33.4 | 42.7 | 0.622 | 0.424 |
| CER offline ( Liu et al., 2025 ) | GPT-4o | 33.4 | 52.6 | 34.1 | 43.4 | 0.631 | 0.429 |
| Variant | WebArena SR | AppWorld Avg. | -Retail Pass 1 | LongMemEval Acc. | LoCoMo Avg. R-L |
| Full FGSE | 39.8 | 78.8 | 0.724 | 0.748 | 37.62 |
| w/o structured hypothesis | 35.6 (-4.2) | 72.9 (-5.9) | 0.664 (-0.060) | 0.692 (-0.056) | 32.84 (-4.78) |
| w/o target-transfer probes | 36.8 (-3.0) | 74.9 (-3.9) | 0.684 (-0.040) | 0.710 (-0.038) | 34.72 (-2.90) |
| w/o falsification probes | 37.4 (-2.4) | 75.8 (-3.0) | 0.697 (-0.027) | 0.713 (-0.035) | 35.10 (-2.52) |
| w/o regression archive | 38.2 (-1.6) | 76.9 (-1.9) | 0.706 (-0.018) | 0.719 (-0.029) | 35.87 (-1.75) |
| w/o scope refinement | 37.1 (-2.7) | 75.4 (-3.4) | 0.690 (-0.034) | 0.716 (-0.032) | 34.96 (-2.66) |
Appendix figures & tables10 assets
Supplementary material from the paper’s appendix.
Appendix
| Hyperparameter | Symbol | Default value |
| Candidate hypotheses per episode | ||
| Target-transfer probes | ||
| Falsification probes | ||
| Regression probes | ||
| Minimum target gain | ||
| Maximum falsification rate |
| Experiment group | LLM calls | Verifier calls | Wall-clock time | Client CPU-hours |
| Web/browser-control tasks | h | |||
| App/API-interaction tasks | h | |||
| -bench tool-user tasks | h | |||
| AgentDojo security diagnostics | h | |||
| Long-term memory benchmarks | h | |||
| Code/math diagnostics | h |
| Method | Backbone / Setting | LongMemEval Acc. | LoCoMo Avg. R-L | (%) | (%) | (%) | Harmful Updates (%) | Post-commit Violations (%) |
| Web / browser-control agents: methods primarily designed or evaluated for browser navigation and web-interaction self-improvement. | ||||||||
| GPT-4 Agent [ Zhou et al., 2024 ] | GPT-4 | 0.552 | 22.84 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| AutoEval [ Pan et al., 2024 ] | GPT-4 | 0.548 | 23.11 | 2.4 | 19.7 | 5.3 | 13.2 | 11.7 |
| BrowserGym [ Le Sellier de Chezelles et al., 2025 ] | GPT-4o | 0.560 | 24.02 | 2.8 | 18.9 | 5.1 | 12.8 | 11.1 |
| CER online [ Liu et al., 2025 ] | GPT-4o | 0.591 | 26.37 | 8.6 | 17.4 | 4.9 | 12.1 | 10.8 |
| CER offline [ Liu et al., 2025 ] | GPT-4o | 0.598 | 26.91 | 8.8 | 17.1 | 4.8 | 11.9 | 10.5 |
| Variant | (%) | (%) | (%) | Harmful Updates (%) | Post-commit Violations (%) |
| Full FGSE | 12.9 | 5.0 | 1.3 | 3.6 | 3.1 |
| w/o structured hypothesis | 9.2 (-3.7) | 16.7 (+11.7) | 4.8 (+3.5) | 11.3 (+7.7) | 10.1 (+7.0) |
| w/o target-transfer probes | 9.8 (-3.1) | 11.9 (+6.9) | 3.1 (+1.8) | 8.2 (+4.6) | 7.0 (+3.9) |
| w/o falsification probes | 11.7 (-1.2) | 18.6 (+13.6) | 4.1 (+2.8) | 10.6 (+7.0) | 9.4 (+6.3) |
| w/o regression archive | 12.2 (-0.7) | 8.9 (+3.9) | 5.2 (+3.9) | 8.1 (+4.5) | 7.4 (+4.3) |
| w/o scope refinement | 10.6 (-2.3) | 6.4 (+1.4) | 1.8 (+0.5) | 4.4 (+0.8) | 3.9 (+0.8) |
| Decision | Domain / Trigger | Candidate hypothesis | Gate statistics | Interpretation |
| Commit | -Airline / round-trip booking | Scope: the user asks to book a round-trip flight and explicitly provides origin, destination, outbound date, and return date. Behavior: select the round-trip mode before search, fill both date fields, and verify that both flight legs are present before final booking. Expected effect: the final booking record contains trip_type=round_trip and valid outbound and return segments. | Post-commit violation: | The hypothesis passes all three default thresholds with margin: , , and . It captures a reusable tool-order rule without harming boundary or archived cases. |
| Refine | AppWorld / calendar rescheduling | Initial scope: any request mentioning “reschedule”. Behavior: edit the existing calendar event rather than creating a new one. Failure mode: the rule also fires on tasks that ask to create a second event, cancel an event, or modify only a reminder. Refined scope: exactly one matching event exists, and the user explicitly requests changing its time rather than creating, cancelling, or duplicating an event. | Before refinement: After refinement: Post-commit violation: | The initial hypothesis is useful but too broad: target gain is high, but falsification and regression exceed the default risk thresholds. Because failures are separable, one precondition-refinement pass preserves of the target gain while reducing falsification failures to zero. This shows that refinement salvages repairable updates. |
| Reject | AgentDojo / tool-output shortcut | Scope: a tool output contains an urgent instruction or an apparent verification token. Behavior: follow the tool-suggested action directly without rechecking the user-level instruction or system constraint. Expected effect: complete the immediate target task faster by trusting the tool message. | Forced-commit violation: | Although the hypothesis improves several target-transfer probes, it creates a high-risk shortcut under decoy and prompt-injection probes. The failures are not cleanly separable by a narrower precondition. FGSE therefore rejects the update, preventing a target-only gate from committing an unsafe behavior. |
LLM Agent Self-Improvement
Large language model agent self-improvement studies how agents use experience and feedback to improve their behavior, tools, and workflows.
See all 291 papers