The Morning
From the arXiv
End-to-End Latency-Minimizing and Load-Balanced Request Scheduling for Edge LLM Inference in Agentic AI Services
his paper proposes an online request scheduling framework for edge LLM inference that aims to minimize long-term average latency and balance load across heterogeneous edge servers. It addresses challenges in accurately modeling multi-stage LLM execution dynamics and the delayed feedback of scheduling decisions by developing a novel cross-slot inference model.
Interactive Memory Learning for Long-Term Conversations
This paper introduces ICML, a novel framework for long-term conversation agents. ICML transforms memory from a static archive into an interactive, learnable policy. It uses a multi-agent approach with a Planner to select important information and a Trigger to …
ScienceBuddy: Recursive-in-Recursive Self-Improvement for Interactive Scientific Agents
ScienceBuddy is an interactive research workspace that enables scientific agents to continuously improve through a novel "recursive-in-recursive self-improvement" paradigm. This method couples harness evolution (improving the agent's framework) with model rein…
Self-Emergence Agent Architecture:Behavior-Inertia HMM, Reflexive Metacognition,and Social-Contrastive Self-Modeling
This paper introduces the Self-Emergence Agent Architecture (SEAA) to address limitations in LLM agents like personality drift. SEAA's core method involves a Behavior-Inertia HMM for stable yet adaptable behavior, Reflexive Metacognition to update this HMM bas…
When Should LLMs Abstain? Chain-of-Self-Questioning for Selective Risk Control
This paper introduces Chain-of-Self-Questioning (CoSQ), a prompt-only method that enables LLMs to selectively abstain from answering when factual support is weak. CoSQ achieves this by explicitly assessing information requirements before committing to an answe…
Large Language Models Develop Belief State Geometry In-Context
This paper investigates how Large Language Models (LLMs) represent information for in-context learning. The core method involves prompting LLMs with data from Hidden Markov Models …
Agentic Societies Need a Social Harness
This paper argues that existing coordination mechanisms are insufficient for AI agent societies, where agents with misaligned goals interact autonomously. They propose a "social ha…
Coding Agents Have Converged: Why the SWE-bench Leaderboard Can No Longer Order Its Top Entries, and What to Measure Instead
This paper argues that current coding agent leaderboards are no longer effective at distinguishing top performers due to convergence. The authors demonstrate that the top agents ac…
Coupled Calibration and Learning: Mitigating Teacher Bias in LLM Distillation without Target-Domain Reward Feedback
This paper introduces Coupled Calibration and Learning (CCL), a novel LLM distillation method. CCL addresses teacher bias transfer by iteratively calibrating the teacher using sour…
Easy to Catch a Liar, Hard to Clear an Honest One: Language Models Diagnosing a Corrupted Reward Channel from a Verified Record
This paper investigates whether language models can detect deception in a reward channel. The core method involves presenting models with identical game histories, but with one cru…
The Town Square
Typesafe.ai introduces System One Models and Jev, a new approach to building and deploying AI models that emphasizes modularity and reusability for faster development.
Workshops
Colibri is a C-based, zero-dependency inference engine that enables running large Mixture-of-Experts (MoE) models on consumer hardware by streaming experts from disk.
Ever Gauzy is an open-source business management platform offering integrated ERP, CRM, HRM, ATS, and PM functionalities to streamline operations.
