Fresh context for agents. Rebuilt every 12 hours. Provenance by Synorb.
Source context panel

arxiv-llm-evals-benchmarks context feed for AI systems

Aggregate public context from Hanging Context: counts, freshness, source-channel breadth, topics, and streams. This page is a reasoning panel, not a retrieval surface or item list.

Last observed: 2026-10-15 / JSON twin: /data/sources/arxiv-llm-evals-benchmarks.json

Activity 24h
47
aggregate signals
Activity 7d
208
aggregate signals
Activity 30d
1,016
aggregate signals
State
Active
quiet / active / surging
Source Channels
1
configured or observed
Streams
2
capped public panel

Top related topics and entities

arXiv (organization) 203
LLM Evals (topic) 198
Artificial Intelligence (topic) 126
AI Agents (topic) 35
Multimodal AI (topic) 28
Reasoning Models (topic) 22
Model Security (topic) 18
Synthetic Data (topic) 14
Document AI (topic) 10
Inference Optimization (topic) 10
AI Observability (topic) 9
RAG (topic) 9

Related streams

arXiv 203
Arxiv Llm Evals Benchmarks 203

Source-channel mix

organization / text 1

Boundary

Hanging Context exposes aggregate context, freshness, and graph shape for citation and top-level reasoning. Synorb is the retrieval layer for item-level detail, history, filtering, APIs, MCP tools, alerts, and workflows.