Skip to content
AI Atlas
Research

Papers

Publications linked to models, labs and benchmarks. Authors, venues and abstracts come from arXiv and publisher pages; model links come from model cards citing the paper.

1,245 papers

Papers
TitleAuthorsOrganizationPublishedIntroduces Models (and artifacts) whose model card or documentation cites this paper — inbound described_by relations.DatasetsBenchmarksCode
arXiv:2110.00061———Modeltable-transformer-detectionModeltable-transformer-structure-recognition———
arXiv:1912.07875———Modelwavlm-base-plusModelwavlm-large———
arXiv:2106.06909———Modelwavlm-base-plusModelwavlm-large———
arXiv:2101.00390———Modelwavlm-base-plusModelwavlm-large———
arXiv:2110.13900———Modelwavlm-base-plusModelwavlm-large———
arXiv:1512.03385———Modelresnet-50———
arXiv:2107.07653———Modeltapex-base-finetuned-wikisql———
arXiv:2601.18184———ModelVibeVoice-ASR———
arXiv:2412.08905———ModelPhi 4———
arXiv:2404.14219———ModelPhi-3.5-vision-instruct———
arXiv:2307.09288———Modelstablelm-3b-4e1tModelLlama-2-13b-chat-hfModelLlama-2-7b-chat-hf+1———
arXiv:2504.07164———ModelMiniMax M2———
arXiv:2509.06501———ModelMiniMax M2.1ModelMiniMax M2———
arXiv:2509.13160———ModelMiniMax M2———
arXiv:2606.13392———ModelMiniMax-M3ArtifactMiniMax-M3-MXFP8———
arXiv:2501.08313———ModelMiniMax-01ModelMiniMax-Text-01-hfModelMiniMax-VL-01———
arXiv:2506.13585———ModelMiniMax-M1-40k-hfModelMiniMax-M1-80k-hfModelMiniMax-M1-80k+1———
arXiv:2505.19641———ModelSynLogic-Mix-3-32BModelSynLogic-32BModelSynLogic-7B———
arXiv:2512.13687———ModelVTP-Small-f16d64ModelVTP-Large-f16d64———
arXiv:2602.11298———ModelVoxtral-Mini-4B-Realtime-2602———
arXiv:2310.06825———ModelSFR-Embedding-MistralModelMistral-7B-Instruct-v0.1ModelMistral-7B-v0.1+1———
arXiv:2507.13264———ModelVoxtral Small 24B 2507ModelVoxtral-Mini-3B-2507———
arXiv:2601.08584———ModelMinistral-3-14B-Reasoning-2512ModelMinistral-3-3B-Reasoning-2512ModelMinistral 3 3B 2512+1———
arXiv:2501.19399———ModelDevstral-Small-2-24B-Instruct-2512———
arXiv:2602.02276———ModelKimi K2.5ModelKimi K2.6———

Author lists and categories are copied from the paper's own metadata (arXiv, publisher). Linked models, datasets, benchmarks and code come from stated relations only; a dash means no source stated one.