ToolGraph enhances multi-turn tool-using agents by integrating schema topology, transition weights, and history-aware controls. Training with DPO on 161 divergence-point preference pairs improves performance: ToolGraph+DPO achieves a 16.8% relative reward gain over baseline, especially in airline and retail tasks, with reward positivity emerging as the key diagnostic signal.
arxiv
arXiv cs.CL
·
47d ago
·
src: 48d ago
·
research
Self-Evolution of Tool-Calling Agents via Divergence-Point Preference Learning
from English
Importance 2/3
New harness with differentiators
arXiv cs.CL
Allen AI
AI agents
Evaluation & benchmarks
Reasoning models
Related articles
arxiv
arXiv cs.CL
·
51d ago
·
1 view
H-RePlan: Hierarchical Recovery for Cross-Device Agent Systems
H-RePlan introduces a hierarchical replanning framework that separates device-local strategy recovery from global orchestrator replanning. It outperforms existing baselines by achieving higher completion and instruction adherence, with reduced token cost, through scope-aware recovery in multi-device agent systems.
arxiv
arXiv cs.LG
·
52d ago
Act2Answer Evaluates Knowledge Retention in Vision-Language-Action Models
Act2Answer introduces a lightweight protocol to assess commonsense and world knowledge retention in VLA models by requiring agents to answer questions through object placement actions. A large-scale study of 7 VLA models and 9 VLM baselines reveals that VLAs perform well on simple concepts but show larger gaps on rich semantic categories compared to their source VLMs, with VQA co-training improving knowledge retention and peak answer-relevant signals observed in middle VLA layers.
lab
Hugging Face Blog
·
25d ago
Skylight's Shippy agent uses deterministic CLI and isolated sandboxing for reliable maritime queries
The Skylight team at Ai2 has built Shippy, an AI agent for real-time maritime domain awareness that prioritizes reliability through a layered architecture of skills, system prompts, and configuration. To ensure correctness when querying live data, Shippy interacts with the Skylight API via a purpose-built CLI rather than constructing raw calls, which eliminates subtle bugs related to pagination and geometry encoding.
arxiv
arXiv cs.LG
·
46d ago
·
1 view
LDT-FRL Framework for Cyber-Resilient IoMT
The LDT-FRL framework introduces a privacy-preserving defense system for IoMT devices, combining temporal attention, lightweight digital twins, and federated reinforcement learning. It achieves 99.66% and 99.95% accuracy on CICDDoS 2019 and TON-IoT benchmarks, with perfect F1 on the MITM class, converging 81% faster than prior methods and offering interpretable defense decisions via SHAP and Grad-CAM.
arxiv
arXiv cs.CL
·
46d ago
·
1 view
MedBench v5: Dynamic Benchmark for Clinical AI
MedBench v5 introduces a dynamic, process-oriented benchmark for clinical multimodal models, featuring clinical cognitive responsiveness and atomic skills across 63 tasks. It includes stressors for degradation analysis and monitors hallucination propagation through five reasoning nodes, revealing that strong task performance does not ensure process stability.