2026-07-02

Internal Generation Record

Internal generation metadata: 453 candidate papers.

Published 2026-07-02 Target source 2026-06-30 Actual source 2026-06-30 Candidates 453 Featured 6 Tracked 20

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 2026-07-01T22:38:26.173615+00:00. Generated at 2026-07-01T22:39:42.587577+00:00. Machine-readable details stay under data/processed and data/reports.

Selected papers

RankTakeawayTopicarXiv
1strengthen multimodal understanding of charts, documents, and visual evidenceTraining and Post-training2606.31732
2improve model reasoning, planning, and verificationTraining and Post-training2606.31645
3make agents use tools and reusable skills more reliablyBenchmarks and Evaluation2606.31179
4make RAG retrieval and knowledge-base QA more reliableBenchmarks and Evaluation2606.31127
5strengthen multimodal understanding of charts, documents, and visual evidenceVision and Image Generation2606.32012
6improve code generation, execution feedback, and automated repairCode Intelligence2606.31993
7make agents use tools and reusable skills more reliablyBenchmarks and Evaluation2606.31966
8improve model reasoning, planning, and verificationBenchmarks and Evaluation2606.31800
9make RAG retrieval and knowledge-base QA more reliableInterpretability2606.31742
10improve image generation, visual understanding, and controllable renderingBenchmarks and Evaluation2606.31700
11improve code generation, execution feedback, and automated repairVision and Image Generation2606.31603
12make agents use tools and reusable skills more reliablyBenchmarks and Evaluation2606.31461
13improve code generation, execution feedback, and automated repairVideo Generation2606.31198
14improve code generation, execution feedback, and automated repairCode Intelligence2606.31082
15strengthen multimodal understanding of charts, documents, and visual evidenceMultimodal Models2606.31054
16make agents use tools and reusable skills more reliablyAgents and Tool Use2606.31023
17make RAG retrieval and knowledge-base QA more reliableVision and Image Generation2606.32039
18strengthen multimodal understanding of charts, documents, and visual evidenceSafety and Alignment2606.31876
19make agents use tools and reusable skills more reliablyCode Intelligence2606.31467
20make RAG retrieval and knowledge-base QA more reliableRetrieval and RAG2606.31184
21improve model reasoning, planning, and verificationBenchmarks and Evaluation2606.31169
22improve code generation, execution feedback, and automated repairCode Intelligence2606.31159
23make agents use tools and reusable skills more reliablyBenchmarks and Evaluation2606.31154
24make agents use tools and reusable skills more reliablyRobotics and Embodied AI2606.31144
25improve code generation, execution feedback, and automated repairSystems and Deployment2606.32020
26strengthen multimodal understanding of charts, documents, and visual evidenceMultimodal Models2606.32016