PRECEPT: Planning Resilience via Experience, Context Engineering & Probing Trajectories A Unified Framework for Test-Time Adaptation with Compositional Rule Learning and Pareto-Guided Prompt Evolution

TL;DR

PRECEPT is a comprehensive framework that enhances large language model agents' test-time adaptation through precise rule retrieval, conflict-aware memory, and Pareto-guided prompt evolution, significantly improving robustness and generalization.

Contribution

It introduces a unified approach combining deterministic rule retrieval, conflict-aware memory, and prompt evolution for improved test-time adaptation in LLM agents.

Findings

Achieves a +41.1pp first-try advantage over previous methods.

Demonstrates 33.3pp improvement in compositional generalization.

Attains 100% success on 2-way logistics compositions.

Abstract

LLM agents that store knowledge as natural language suffer steep retrieval degradation as condition count grows, often struggle to compose learned rules reliably, and typically lack explicit mechanisms to detect stale or adversarial knowledge. We introduce PRECEPT, a unified framework for test-time adaptation with three tightly coupled components: (1) deterministic exact-match rule retrieval over structured condition keys, (2) conflict-aware memory with Bayesian source reliability and threshold-based rule invalidation, and (3) COMPASS, a Pareto-guided prompt-evolution outer loop. Exact retrieval eliminates partial-match interpretation errors on the deterministic path (0% by construction, vs 94.4% under Theorem~B.6's independence model at N=10) and supports compositional stacking through a semantic tier hierarchy; conflict-aware memory resolves static--dynamic disagreements and supports drift adaptation; COMPASS evaluates prompts through the same end-to-end execution pipeline. Results (9--10 seeds): PRECEPT achieves a +41.1pp first-try advantage over Full Reflexion (d>1.9), +33.3pp compositional generalization (d=1.55), 100% $P_1$ on 2-way logistics compositions (d=2.64), +40--55pp continuous learning gains, strong eventual robustness under adversarial static knowledge (100% logistics with adversarial SK active; partial recovery on integration), +55.0pp drift recovery (d=0.95, p=0.031), and 61% fewer steps. Core comparisons are statistically significant, often at p<0.001.