CONTRAMEM: Learning Self-Evolving Procedural Memory from Contrasting Multi-Model Trajectories
cs.AI
Submitted: 2026-08-23
Updated: 2026-09-30
Comments: 35 pages, 7 figures; includes technical appendix
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- ARE: Scaling Up Agent Environments and Evaluations
- MemCollab: Cross-Model Memory Collaboration via Contrastive Trajectory Distillation
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Trajectory-Informed Memory Generation for Self-Improving Agent Systems
- Gaia2: Benchmarking LLM Agents on Dynamic and Asynchronous Environments
- Momentum Contrast for Unsupervised Visual Representation Learning
- Muon-OGD: Muon-based Spectral Orthogonal Gradient Projection for LLM Continual Learning
- ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory
- MemGPT: Towards LLMs as Operating Systems
- Agent KB: Leveraging Cross-Domain Experience for Agentic Problem Solving
- Representation Learning with Contrastive Predictive Coding
- Voyager: An Open-Ended Embodied Agent with Large Language Models
- Agent Workflow Memory
- How Memory Management Impacts LLM Agents: An Empirical Study of Experience-Following Behavior
- Useful Memories Become Faulty When Continuously Updated by LLMs
- Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems