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
Title	Authors	Organization	Published	Introduces	Datasets	Benchmarks	Code
arXiv:2506.20923	—	—	—	ModelKaLM-Embedding-Gemma3-12B-2511	—	—	—
arXiv:2501.01028	—	—	—	ModelKaLM-Embedding-Gemma3-12B-2511	—	—	—
arXiv:2608.24053	—	—	—	ModelWeMM-Embedding-2B	—	—	—
arXiv:2607.04884	—	—	—	ModelHunyuanOCR	—	—	—
arXiv:2409.02095	—	—	—	ModelDepthCrafter	—	—	—
arXiv:2512.24618	—	—	—	ModelYoutu-LLM-2B	—	—	—
arXiv:2602.06932	—	—	—	Artifacttogethercomputer/Aurora-Spec-Qwen3-Coder-Next-FP8ModelAurora-Spec-Minimax-M2.5	—	—	—
arXiv:2302.10866	—	—	—	ModelStripedHyena-Hessian-7BModelStripedHyena-Nous-7BModelevo-1-131k-base+1	—	—	—
arXiv:2203.14343	—	—	—	Modelevo-1-131k-baseModelevo-1-8k-base	—	—	—
arXiv:2310.18780	—	—	—	ModelStripedHyena-Hessian-7BModelStripedHyena-Nous-7BModelevo-1-131k-base+1	—	—	—
arXiv:2206.11893	—	—	—	Modelevo-1-131k-baseModelevo-1-8k-base	—	—	—
arXiv:2303.06349	—	—	—	Modelevo-1-131k-baseModelevo-1-8k-base	—	—	—
arXiv:2102.02611	—	—	—	Modelevo-1-131k-baseModelevo-1-8k-base	—	—	—
arXiv:2210.09298	—	—	—	Modelevo-1-131k-baseModelevo-1-8k-base	—	—	—
arXiv:2310.12109	—	—	—	Modelm2-bert-80M-2k-retrievalModelm2-bert-80M-32k-retrievalModelm2-bert-80M-8k-retrieval	—	—	—
arXiv:2307.03172	—	—	—	Modeltogethercomputer/Llama-2-7B-32K-Instruct	—	—	—
arXiv:2504.10449	—	—	—	ModelM1-3B	—	—	—
arXiv:2311.05908	—	—	—	ModelStripedHyena-Hessian-7BModelStripedHyena-Nous-7B	—	—	—
arXiv:2205.05131	—	—	—	ModelGPT-JT-6B-v1	—	—	—
arXiv:2210.11399	—	—	—	ModelGPT-JT-6B-v1	—	—	—
arXiv:2602.15763	—	—	—	ModelGLM 5ModelGLM 5.1Artifactzai-org/GLM-5.1-FP8+5	—	—	—
arXiv:2103.10360	—	—	—	Modelchatglm3-6bModelchatglm2-6b	—	—	—
arXiv:2210.02414	—	—	—	Modelchatglm3-6bModelchatglm2-6b	—	—	—
arXiv:1911.02150	—	—	—	Modelchatglm2-6b	—	—	—
arXiv:2406.12793	—	—	—	Modelchatglm3-6bModelglm-4-9b-chatModelchatglm2-6b	—	—	—

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.