Agents

LLM agents and the systems around them: harnesses and orchestration, tool use and MCP, multi-agent systems, agent memory and planning

600 articles · page 13 of 13

Agents

Selecting complementary LLM skills improves success while reducing context use

Chen et al. formulate the choice of reusable skill documents for LLM agents as a budget-constrained optimization problem rather than an independent relevance-ranking task. Their Best Prefix Selection algorithm achieved 0.73 task success on a contamination-controlled BigCodeBench variant, compared with 0.20–0.52 for existing selection methods, while using fewer tokens than the strongest released router.

Agents

New framework lets you train AI agents inside the same harness systems they use at inference

The authors present OpenForgeRL, an open-source framework that trains stateful, harness-based AI agents end-to-end using standard reinforcement learning libraries. It wraps the harness's model calls via a proxy for data recording and orchestrates rollouts in remote Kubernetes containers, yielding competitive results on tool-use, browser, and computer-use benchmarks with relatively few training tasks.

Agents

Python objects become AI agents in new NVIDIA framework

The authors introduce NVIDIA Object-Oriented Agents (NOOA), a model-agnostic Python framework where any Python object can act as an AI agent: methods define available actions, fields store state, docstrings serve as prompts, and type annotations act as contracts. Methods whose body consists only of '...' are completed at runtime by an LLM-driven loop, while normal methods remain deterministic. The framework is demonstrated on SWE-bench Verified, Terminal-Bench 2.0, and ARC-AGI-3, showing that current models use this interface effectively.

Agents

Re-evaluation shows harness evolution for agents may not outperform simple test-time scaling

Wang et al. revisit the evaluation of automatic harness evolution for LLM agents, comparing it with simple test-time scaling and discovery baselines under matched feedback and inference budgets. They find that harness evolution does not consistently outperform these baselines on Terminal-Bench 2.1 and shows limited generalization to held-out tasks, raising concerns about overfitting and the effectiveness of the approach.

Agents

Treating memory management as a trainable cognitive skill boosts LLM performance 2x–4x in long-horizon tasks

The authors introduce AutoMem, a framework that treats memory management as a learnable skill for LLMs by promoting file-system operations to first-class actions. By automating the optimization of memory structure (prompts, schemas, action vocabulary) and model proficiency (training on good memory decisions), AutoMem improved base agent performance by approximately 2x to 4x across three long-horizon games without altering task-action behavior.

Agents

Three-stage training instills world model planning into LLM agents

Zhang et al. propose a training paradigm that equips LLM agents with an internal world model for prospective reasoning. By first injecting predictive capabilities through a world model mid-training stage, then eliciting structured foresight via supervised fine-tuning, and finally calibrating with reinforcement learning, the authors demonstrate significant improvements on search and mathematical reasoning tasks over standard fine-tuning approaches.

Agents

Latent communication across different LLMs matches or beats text-based transfer at lower compute.

The authors investigate whether heterogeneous LLMs can be aligned to directly transfer latent representations (KV-caches) between them, bypassing costly text decoding and re-encoding. They propose a method using a cross-model transformation and two-phase training (reconstruction then generation), achieving performance comparable to or better than text communication in context-aware settings at 2-3x lower compute, and remaining effective in context-unaware settings where prior heterogeneous methods collapsed.

Agents

Decentralized agents with shared context outperform centralized orchestration

The authors propose Decentralized Language Models (DeLM), a multi-agent framework that replaces a central controller with a shared context and task queue, allowing agents to asynchronously claim subtasks and build on verified progress. On SWE-bench Verified, DeLM improved performance by up to 10.5 percentage points and reduced costs by roughly 50%. On LongBench-v2, it achieved the highest average accuracy across four frontier model families.

Agents

Pruning context to recent tool calls improves agent reliability and efficiency

The authors test four context engineering strategies for GPT-5 agents using Model Context Protocol tools on a 50-task hotel expense benchmark. They find that pruning context to the last five tool call/response pairs and adding automated summarization yields the best results: 91.6% complete itemization, 99.64% amount itemized, with 63% fewer tokens and 60% less runtime than retaining full conversation history.

Agents

VLMs improve spatial reasoning by actively imagining novel views with a world simulator

The authors propose Astra, an agentic spatial reasoning framework that combines two components: a world simulator (Astra-WM) that generates novel-view images from context images and natural-language camera motions, and an RL-trained VLM policy (Astra-VL) that decides when to query the simulator. Trained with view consistency tuning and a two-phase RL curriculum, Astra improves spatial reasoning benchmarks by 9–10 points over direct VLM answering, demonstrating that effective world-model-augmented reasoning requires learning when, where, and how to imagine.

Agents

Self-supervised method improves agent harnesses using past trajectories

The authors introduce Retrospective Harness Optimization (RHO), a self-supervised method that optimizes an AI agent's harness of skills, tools, and workflows using only past task trajectories. By selecting a diverse coreset of challenging tasks, re-solving them in parallel, and using self-validation to pick the best harness update, the method improves pass rate on SWE-Bench Pro from 59% to 78% in a single optimization round.

Agents

Proposing a workflow store to harden AI agents against failure

The authors critique the dominant on-the-fly paradigm for AI agents, which synthesizes plans and executes actions rapidly in response to prompts, arguing it bypasses established software engineering processes like testing and adversarial evaluation. They propose an AI Workflow Store of hardened, reusable workflows to amortize the cost of rigor across users, aiming to improve reliability and security for high-stakes applications.