The Morning
From the arXiv
Cognitive-structured Multimodal Agent for Multimodal Understanding, Generation, and Editing
his paper introduces a Cognitive-structured Multimodal Agent that addresses limitations in current unified multimodal models by externalizing visual information into an Episodic Visual Memory. This allows for selective retrieval of relevant visual context during reasoning, overcoming issues with long-horizon dialogues and token explosion. The agent's contribution lies in its novel architecture with distinct engines for perception, memory retrieval, and task execution, along with a new dataset generation method to facilitate training for improved cross-turn multimodal understanding and generation.
Game Theory Driven Multi-Agent Framework Mitigates Language Model Hallucination
This paper introduces G-Frame, a multi-agent framework that uses game theory to train language models for scientific domains. By integrating Bayesian and team game principles, G-Frame forces models to internalize domain constraints, significantly reducing hall…
Towards Mechanistically Understanding Why Memorized Knowledge Fails to Generalize in Large Language Model Finetuning
This paper addresses the "Knowing-Using Gap" in LLM fine-tuning, where models memorize new facts but fail to apply them in reasoning. Using a novel "self-patching" technique, they identify that memorized knowledge is internally present but not correctly routed…
UltraX: Refining Pre-Training Data at Scale with Adaptive Programmatic Editing
UltraX addresses the diminishing returns of scaling LLMs by focusing on data quality. Its core method is an adaptive programmatic editing framework that refines pre-training data at scale by enabling fine-grained insertion, deletion, and modification. This app…
WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search
WebSwarm addresses the limitations of single LLM agents in deep and wide web searches by introducing a recursive delegation framework. Its core method involves dynamically creating agentic search nodes that jointly decompose tasks, recursively expand search, a…
Multi-Modal, Multi-Environment Machine Teaching for Robust Reward Learning
This paper proposes a novel machine teaching method for robust reward learning in Inverse Reinforcement Learning (IRL). It addresses the challenge of learning reward functions that…
Remember When It Matters: Proactive Memory Agent for Long-Horizon Agents
This paper introduces a proactive memory agent that actively intervenes by injecting relevant memories into an action agent's decision-making process. This combats "behavioral stat…
SMetric: Rethink LLM Scheduling for Serving Agents with Balanced Session-centric Scheduling
This paper introduces SMetric, a novel LLM scheduling method for serving AI agents. It addresses the unique demands of agentic workloads, where complete responses are prioritized o…
TRACE: A Two-Channel Robust Attribution Watermark via Complementary Embeddings for LLM-Agent Trajectories
TRACE embeds a robust watermark into LLM agent trajectories by using two complementary channels. One channel is content-based for deletion resilience, and the other is position-bas…
WCog-VLA: A Dual-Level World-Cognitive Vision-Language-Action Model for End-to-End Autonomous Driving
WCog-VLA addresses the limitations of reactive autonomous driving by introducing a dual-level framework for proactive driving. It achieves this by unifying world cognition and reas…
The Town Square
Meta's Muse Spark 1.1 is a new API offering a powerful, open-source text-to-image generation model for developers to build creative AI applications.
Workshops
This repository provides production-grade engineering skills for AI coding agents, enabling them to perform complex tasks like code generation, debugging, and testing.
This repository offers a curated collection of DESIGN.md files from popular brand design systems, enabling coding agents to automatically generate matching UIs when dropped into a project.
