仅用于站内搜索，没有排版格式，具体信息请跳转上方微信公众号内链接
Datawhale分享
必读论文：LLM/AI，编辑：深度学习自然语言处理
项目地址：https ://github. com/InterviewReady/ai-engineering-resources
Byte-pairEncodinghttps ://arxiv. org/pdf/1508. 07909
ByteLatentTransformer:PatchesScaleBetterThanTokenshttps ://arxiv. org/pdf/2412. 09871
IMAGEBIND:OneEmbeddingSpaceToBindThemAllhttps ://arxiv. org/pdf/2305. 05665
FAISSlibraryhttps ://arxiv. org/pdf/2401. 08281
FacebookLargeConceptModelshttps ://arxiv. org/pdf/2412. 08821v2
TensorFlowhttps ://arxiv. org/pdf/1605. 08695
Deepseekfilesystemhttps ://github. com/deepseek-ai/3FS/blob/main/docs/design_notes. md
MilvusDBhttps ://www. cs.purdue. edu/homes/csjgwang/pubs/SIGMOD21_Milvus. pdf
BillionScaleSimilaritySearch:FAISShttps ://arxiv. org/pdf/1702. 08734
Rayhttps ://arxiv. org/abs/1712. 05889
AttentionisAllYouNeedhttps ://papers. neurips.cc/paper/7181-attention-is-all-you-need. pdf
FlashAttentionhttps ://arxiv. org/pdf/2205. 14135
MultiQueryAttentionhttps ://arxiv. org/pdf/1911. 02150
GroupedQueryAttentionhttps ://arxiv. org/pdf/2305. 13245
GoogleTitansoutperformTransformershttps ://arxiv. org/pdf/2501. 00663
VideoRoPE:RotaryPositionEmbeddinghttps ://arxiv. org/pdf/2502. 05173
Sparsely-GatedMixture-of-ExpertsLayerhttps ://arxiv. org/pdf/1701. 06538
GShardhttps ://arxiv. org/abs/2006. 16668
SwitchTransformershttps ://arxiv. org/abs/2101. 03961
DeepReinforcementLearningwithHumanFeedbackhttps ://arxiv. org/pdf/1706. 03741
Fine-TuningLanguageModelswithRHLFhttps ://arxiv. org/pdf/1909. 08593
TraininglanguagemodelswithRHLFhttps ://arxiv. org/pdf/2203. 02155
Chain-of-ThoughtPromptingElicitsReasoninginLargeLanguageModelshttps ://arxiv. org/pdf/2201. 11903
Chainofthoughthttps ://arxiv. org/pdf/2411. 14405v1/
DemystifyingLongChain-of-ThoughtReasoninginLLMshttps ://arxiv. org/pdf/2502. 03373
TransformerReasoningCapabilitieshttps ://arxiv. org/pdf/2405. 18512
Scalemodeltesttimesisbetterthanscalingparametershttps ://arxiv. org/pdf/2408. 03314
TrainingLargeLanguageModelstoReasoninaContinuousLatentSpacehttps ://arxiv. org/pdf/2412. 06769
DeepSeekR1https ://arxiv. org/pdf/2501. 12948v1
LatentReasoning:ARecurrentDepthApproachhttps ://arxiv. org/pdf/2502. 05171
TheEraof1-bitLLMs:AllLargeLanguageModelsarein1. 58Bitshttps ://arxiv. org/pdf/2402. 17764
ByteDance1. 58https ://arxiv. org/pdf/2412. 18653v1
TransformerSquarehttps ://arxiv. org/pdf/2501. 06252
1boutperforms405bhttps ://arxiv. org/pdf/2502. 06703
SpeculativeDecodinghttps ://arxiv. org/pdf/2211. 17192
DistillingtheKnowledgeinaNeuralNetworkhttps ://arxiv. org/pdf/1503. 02531
BYOL-DistilledArchitecturehttps ://arxiv. org/pdf/2006. 07733
DINOhttps ://arxiv. org/pdf/2104. 14294
RWKV:ReinventingRNNsfortheTransformerErahttps ://arxiv. org/pdf/2305. 13048
Mambahttps ://arxiv. org/pdf/2312. 00752
DistillingTransformerstoSSMshttps ://arxiv. org/pdf/2408. 10189
LoLCATs:OnLow-RankLinearizingofLargeLanguageModelshttps ://arxiv. org/pdf/2410. 10254
ThinkSlow,Fasthttps ://arxiv. org/pdf/2502. 20339
GoogleMathOlympiad2https ://arxiv. org/pdf/2502. 03544
CompetitiveProgrammingwithLargeReasoningModelshttps ://arxiv. org/pdf/2502. 06807
GoogleMathOlympiad1https ://www. nature.com/articles/s41586-023-06747-5
CanAIbemadetothinkcriticallyhttps ://arxiv. org/pdf/2501. 04682
EvolvingDeeperLLMThinkinghttps ://arxiv. org/pdf/2501. 09891
LLMsCanEasilyLearntoReasonfromDemonstrationsStructurehttps ://arxiv. org/pdf/2502. 07374
Separatingcommunicationfromintelligencehttps ://arxiv. org/pdf/2301. 06627
Languageisnotintelligencehttps ://gwern. net/doc/psychology/linguistics/2024-fedorenko. pdf
Imageis16x16wordhttps ://arxiv. org/pdf/2010. 11929
CLIPhttps ://arxiv. org/pdf/2103. 00020
deepseekimagegenerationhttps ://arxiv. org/pdf/2501. 17811
ViViT:AVideoVisionTransformerhttps ://arxiv. org/pdf/2103. 15691
JointEmbeddingabstractionswithself-supervisedvideomaskshttps ://arxiv. org/pdf/2404. 08471
FacebookVideoJAMaigenhttps ://arxiv. org/pdf/2502. 02492
AutomatedUnitTestImprovementusingLargeLanguageModelsatMetahttps ://arxiv. org/pdf/2402. 09171
OpenAIo1SystemCardhttps ://arxiv. org/pdf/2412. 16720
LLM-poweredbugcatchershttps ://arxiv. org/pdf/2501. 12862
Chain-of-RetrievalAugmentedGenerationhttps ://arxiv. org/pdf/2501. 14342
SwarmbyOpenAIhttps ://github. com/openai/swarm
ModelContextProtocolhttps ://www. anthropic.com/news/model-context-protocol
uberqueryGPThttps ://www. uber.com/en-IN/blog/query-gpt/
