The Morning
From the arXiv
Closing the Consistency Gap: Self-Evolving Agents That Learn to Stay on Course
his paper addresses the "consistency gap" in LLM agents, where agents are often accurate on average but unreliable across multiple runs. Their core method involves a "self-evolving agent framework" that identifies unstable steps in agent reasoning and converts them into episodic memory. This memory then guides future executions, significantly reducing the inconsistency gap and improving trustworthiness.
Measuring LLM Sycophancy under Sustained Multi-Turn Pressure
This paper introduces SPINE, a benchmark designed to measure LLM sycophancy under prolonged, adaptive disagreement. Unlike previous evaluations, SPINE uses a persistent LLM proxy to challenge the target model for up to 25 turns. The study reveals that sycophan…
MeClear: Cooperative Game-Theoretic Attribution and Risk-Aware Memory Clearance for Long-Horizon LLM Agents
MeClear is a novel framework for long-horizon LLM agents that addresses the problem of outdated or conflicting information in their external memory. It uses cooperative game-theoretic attribution to identify memories with negative downstream utility and select…
Omni Interaction Agent Technical Report
This paper introduces Gander, an end-to-end model that unifies multi-modal perception, real-time interaction, and agentic capabilities. Its core method is a Cerebellum-Brain collaborative framework, enabling continuous, full-duplex interaction where the Cerebe…
Procedural Graphs: Self-Evolving Execution Structures for LLM Agents
This paper introduces the Procedural Graph, a novel structure that organizes procedural knowledge for LLM agents. Unlike traditional methods, it uses (procedure, relation, procedure) triplets to guide action selection. The graph self-evolves, allowing agents t…
Training-Free Task Vectors for LLM Behavioral Control
This paper introduces Training-Free Task Vectors (TFTVs), a method to control LLM behavior without costly fine-tuning. TFTVs compute task-specific directions in model weights using…
Transformers as In-Context Samplers: From Closed-Form Diffusion to Estimation-Free Sampling
This paper demonstrates that frozen transformers can perform data generation by simulating iterative samplers from in-context examples. The core method involves constructing transf…
AuK Technical Report: An Open-Source Foundational Model for Speech Generation and Editing
AuK is an open-source foundational model that unifies speech generation and editing using natural language instructions and audio context. Its core method involves a multimodal LLM…
Experience Funnel: A State-Policy Alternating Loop for Self-Evolving Agents
The Experience Funnel framework enables self-evolving agents by alternating between fast textual state updates and slow policy consolidation. This allows agents to rapidly adapt to…
Co-Evolving Harnesses and Models: On-Policy Correction Helps Weaker Models Catch Up Where Imitation Fails
This paper introduces a method for co-evolving agent harnesses (the surrounding system components) and models. They found that evolving the harness alongside a weaker model, and th…
The Town Square
LibreOffice achieved record downloads after announcing it has no AI features, suggesting a user preference for traditional, non-AI software.
Workshops
This repository provides 38 self-contained HTML+SVG editorial diagram types for Claude Code, Codex, and Pi, designed for clean, shadow-free rendering without Mermaid.
The OpenAI Skills repository serves as a catalog for Codex, providing a collection of reusable code snippets and functionalities to enhance AI-powered code generation and understanding.
