ParamMem: Augmenting Language Agents with Parametric Reflective Memory

TL;DR

ParamMem introduces a parametric memory module that enhances language agents by increasing reflective diversity, leading to improved reasoning and problem-solving across multiple tasks with efficient, scalable, and self-improving capabilities.

Contribution

The paper presents ParamMem, a novel parametric memory module that encodes reflection patterns into model parameters, enabling diverse and effective self-reflection in language agents.

Findings

ParamMem improves task success across multiple benchmarks.

ParamMem is sample-efficient and scalable across model sizes.

ParamMem enables self-improvement without external models.

Abstract

Self-reflection enables language agents to iteratively refine solutions, yet often produces repetitive outputs that limit reasoning performance. Recent studies have attempted to address this limitation through various approaches, among which increasing reflective diversity has shown promise. Our empirical analysis reveals a strong positive correlation between reflective diversity and task success, further motivating the need for diverse reflection signals. We introduce ParamMem, a parametric memory module that encodes cross-sample reflection patterns into model parameters, enabling diverse reflection generation through temperature-controlled sampling. Building on this module, we propose ParamAgent, a reflection-based agent framework that integrates parametric memory with episodic and cross-sample memory. Extensive experiments on code generation, mathematical reasoning, and multi-hop question answering demonstrate consistent improvements over state-of-the-art baselines. Further analysis reveals that ParamMem is sample-efficient, enables weak-to-strong transfer across model scales, and supports self-improvement without reliance on stronger external model, highlighting the potential of ParamMem as an effective component for enhancing language agents.