Berkeley AI Research
Jul 26
Teaching LLMs to Update Beliefs for Efficient Long-Horizon Interaction
abbel-fig { display: block; text-align: center; margin: 2. 4em 0; line-height: 1.
AI agents
Tool use, function calling, orchestration and the protocols that let models act rather than only answer.
Towards Data Science
Jul 24
Context Windows Forget What Matters — I Built a Usage-Reinforced Decay Engine for AI Agent Memory
Most AI memory systems keep the newest information—not the most important. Here's how I used the Ebbinghaus forgetting curve to build a better memory engine for LLMs.
arXiv AI
Jul 24
Workload-Aware Caching for Multi-Agent Systems
arXiv:2607. 20495v1 Announce Type: new Abstract: Multi-agent systems decompose complex tasks into directed acyclic graphs (DAGs) of specialized agent executions, creating natural opportunities for caching intermediate results across queries.
arXiv Machine Learning
Jul 24
Approximate Quantum State Preparation Through Proximal Policy Optimization
arXiv:2607. 21121v1 Announce Type: cross Abstract: In this work, a quantum architecture search framework for approximate quantum state preparation (QSP) is proposed.
arXiv AI
Jul 24
GRADRAG: Cross-Component Prompt Adaptation for Coordinated Multi-Agent RAG
arXiv:2607. 21324v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) systems increasingly employ multiple LLM agents.
arXiv AI
Jul 24
LeanFlow: A Case Study in Workflow-Driven Lean Autoformalization
arXiv:2607. 20503v1 Announce Type: new Abstract: We present and evaluate LeanFlow, an LLM agent system specialized for translating mathematical papers into buildable Lean projects.
arXiv AI
Jul 24
CMI-Mem: Toward Generalizable Long-Term Memory Management via CMI-Augmented Reinforcement Learning
arXiv:2607. 20553v1 Announce Type: new Abstract: Memory Manager models are pivotal in agent systems.
arXiv AI
Jul 24
InferenceBench: A Benchmark for Open-Ended LLM Inference Optimization by AI Agents
arXiv:2607. 20468v1 Announce Type: new Abstract: AI agents are increasingly used to automate research and development tasks, yet existing benchmarks typically evaluate them on prescribed workflows or narrow action spaces.
arXiv AI
Jul 24
Evaluating and Guarding Citation Faithfulness in Agentic Scientific Synthesis
arXiv:2607. 20527v1 Announce Type: new Abstract: Agentic LLM systems such as OpenScholar and PaperQA2 read the scientific literature and return cited answers, and both they and their benchmarks already check whether those citations hold, with a fixed attribution model or human graders.
arXiv AI
Jul 24
AINTMA: Agentic AI Architecture for Autonomous Test Management with Generative Intelligence, Secure Cloud Communication and Adaptive Quality Analytics
arXiv:2607. 20452v1 Announce Type: new Abstract: Modern software quality assurance demands intelligent, autonomous systems capable of adaptive decision-making across distributed cloud environments.
arXiv AI
Jul 24
Regulating autonomous and agentic AI
arXiv:2607. 21345v1 Announce Type: new Abstract: Regulating activities where regulatees use autonomous and agentic AI is challenging.
arXiv AI
Jul 24
Verifier-First Evaluation of Agentic LLMs for Infrastructure-as-Code Generation
arXiv:2607. 20478v1 Announce Type: cross Abstract: Infrastructure-as-Code (IaC) generation from natural language requires satisfying provider schemas, dependency planning, and organizational policy constraints, not merely producing syntactically plausible configurations.
arXiv AI
Jul 24
TOUR: A Trajectory-Level Unlearning Benchmark for Offline Reinforcement Learning
arXiv:2607. 21111v1 Announce Type: cross Abstract: Offline Reinforcement Learning (RL) agents are trained on fixed behavioral trajectories, which makes trajectory-level deletion important when selected data must be removed after training.
arXiv AI
Jul 24
AISE-Bench: A Full-Cycle Curated Benchmark for Information Seeking on Academic Knowledge Graphs
arXiv:2607. 20498v1 Announce Type: new Abstract: Large language models (LLMs) augmented with tools are emerging as autonomous agents capable of using Web engine, APIs, and code to solve complex, long-horizon tasks.
arXiv AI
Jul 24
Euclid-MCP: A Model Context Protocol Server for Deterministic Logical Reasoning via Prolog
arXiv:2607. 21412v1 Announce Type: new Abstract: Large Language Models (LLMs) excel at natural language understanding and generation but remain unreliable for multi-step logical reasoning, especially in safety-critical or compliance-sensitive domains.
arXiv AI
Jul 24
LLM-INSTRUCT at UZH Shared Task 2026: Constraint-Aware Retrieval and Selective Debate for Paragraph-Level Argument Mining
arXiv:2607. 20430v1 Announce Type: cross Abstract: We present LLM-INSTRUCT, the winning system for the UZH Shared Task at ArgMining 2026 on paragraph-level argument mining in UN and UNESCO resolutions.
arXiv AI
Jul 24
EmoAgent-R1: Towards Multimodal Emotion Understanding with Reinforcement Learning-based Dynamic Agent Specialization
arXiv:2607. 21013v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have achieved impressive performance in multimodal emotion recognition (MER) tasks and lifted MER to a new level that is complex emotion understanding with advanced video understanding abilities and natural language description.
arXiv AI
Jul 24
MiniCache: Reusable Program Caching with Small Model Interfaces for Efficient LLM Inference
arXiv:2607. 20507v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for program-aided reasoning, agentic decision making, and structured task execution, but these applications often incur high inference cost.
arXiv AI
Jul 24
Adaptive Multi-Horizon Reinforcement Learning
arXiv:2607. 20656v1 Announce Type: cross Abstract: Effective decision-making in complex and changing environments requires balancing short-term and long-term consequences.
arXiv AI
Jul 24
OpenForgeRL: Train Harness-native Agents in Any Environment
arXiv:2607. 21557v1 Announce Type: new Abstract: Modern AI agents rely on elaborate inference harnesses such as Claude Code, Codex, and OpenClaw to drive multi-turn reasoning, tool use, and access to external systems.