2026-09-06

Internal Generation Record

Internal generation metadata: 405 candidate papers.

Published 2026-09-06 Target source 2026-09-04 Actual source 2026-09-03 Candidates 405 Featured 6 Tracked 20 Source date fallback

Generation Record

This page preserves selected papers, candidate scale, and source-date metadata for traceability. The page only changes presentation, not selected papers, ordering, or counts.

Internal generation record. Fetched at 2026-09-05T22:50:13.866134+00:00. Generated at 2026-09-05T22:51:47.996951+00:00. Machine-readable details stay under data/processed and data/reports.

Selected papers

RankTakeawayTopicarXiv
27make agents use tools and reusable skills more reliablyAgents and Tool Use2609.03920
28strengthen multimodal understanding of charts, documents, and visual evidenceBenchmarks and Evaluation2609.03806
29strengthen multimodal understanding of charts, documents, and visual evidenceMultimodal Models2609.03804
30improve model reasoning, planning, and verificationMultimodal Models2609.03756
33strengthen multimodal understanding of charts, documents, and visual evidenceCode Intelligence2609.03721
34make agents use tools and reusable skills more reliablyVideo Generation2609.03673
31make agents use tools and reusable skills more reliablyAgents and Tool Use2609.03753
32make agents use tools and reusable skills more reliablyAgents and Tool Use2609.03727
35make agents use tools and reusable skills more reliablyAgents and Tool Use2609.03590
36make agents use tools and reusable skills more reliablyVision and Image Generation2609.03586
37make agents use tools and reusable skills more reliablyBenchmarks and Evaluation2609.03553
38make agents use tools and reusable skills more reliablyMultimodal Models2609.03544
39improve model reasoning, planning, and verificationCode Intelligence2609.03522
40improve code generation, execution feedback, and automated repairCode Intelligence2609.03516
41make RAG retrieval and knowledge-base QA more reliableVision and Image Generation2609.03505
42make RAG retrieval and knowledge-base QA more reliableBenchmarks and Evaluation2609.03470
43improve code generation, execution feedback, and automated repairBenchmarks and Evaluation2609.03464
44make RAG retrieval and knowledge-base QA more reliableSystems and Deployment2609.03459
45make RAG retrieval and knowledge-base QA more reliableTraining and Post-training2609.03432
46make agents use tools and reusable skills more reliablySpeech and Audio2609.03423
47make agents use tools and reusable skills more reliablyAgents and Tool Use2609.03414
48make agents use tools and reusable skills more reliablyReasoning and Planning2609.03379
49make agents use tools and reusable skills more reliablyAgents and Tool Use2609.03340
50make agents use tools and reusable skills more reliablyBenchmarks and Evaluation2609.03293
51improve code generation, execution feedback, and automated repairBenchmarks and Evaluation2609.03258
52make agents use tools and reusable skills more reliablyAgents and Tool Use2609.03236