2026-07-21

Internal Generation Record

Internal generation metadata: 267 candidate papers.

Published 2026-07-21 Target source 2026-07-19 Actual source 2026-07-17 Candidates 267 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 2026-07-20T22:09:50.733093+00:00. Generated at 2026-07-20T22:11:00.372177+00:00. Machine-readable details stay under data/processed and data/reports.

Selected papers

RankTakeawayTopicarXiv
1improve model reasoning, planning, and verificationReasoning and Planning2607.15589
2make agents use tools and reusable skills more reliablyBenchmarks and Evaluation2607.16156
3make agents use tools and reusable skills more reliablyAgents and Tool Use2607.16038
4make agents use tools and reusable skills more reliablyAgents and Tool Use2607.15781
5make agents use tools and reusable skills more reliablyVideo Generation2607.15699
6make RAG retrieval and knowledge-base QA more reliableTraining and Post-training2607.16031
7improve code generation, execution feedback, and automated repairCode Intelligence2607.15889
8make agents use tools and reusable skills more reliablyAgents and Tool Use2607.15880
9strengthen multimodal understanding of charts, documents, and visual evidenceBenchmarks and Evaluation2607.15851
10strengthen multimodal understanding of charts, documents, and visual evidenceMultimodal Models2607.15779
11improve code generation, execution feedback, and automated repairBenchmarks and Evaluation2607.15619
12make agents use tools and reusable skills more reliablySystems and Deployment2607.15516
13make agents use tools and reusable skills more reliablyAgents and Tool Use2607.16189
14test temporal consistency and motion realism in video generationSystems and Deployment2607.16154
15improve code generation, execution feedback, and automated repairCode Intelligence2607.16127
16make agents use tools and reusable skills more reliablyAgents and Tool Use2607.16066
17make RAG retrieval and knowledge-base QA more reliableCode Intelligence2607.16012
18make agents use tools and reusable skills more reliablyMultimodal Models2607.15942
19make RAG retrieval and knowledge-base QA more reliableTraining and Post-training2607.15928
20make agents use tools and reusable skills more reliablyAgents and Tool Use2607.15901
21improve code generation, execution feedback, and automated repairBenchmarks and Evaluation2607.15899
22improve code generation, execution feedback, and automated repairSystems and Deployment2607.15865
23make RAG retrieval and knowledge-base QA more reliableSystems and Deployment2607.15810
24make agents use tools and reusable skills more reliablyAgents and Tool Use2607.15769
25make agents use tools and reusable skills more reliablyBenchmarks and Evaluation2607.15768
26make RAG retrieval and knowledge-base QA more reliableBenchmarks and Evaluation2607.15766