- 112asynergisticsparseandlowrankcompressionmethodforlargelanguagemodels
- 2columns1rowarussianbenchmarkfortextualandmultimodaltableunderstandingandreasoning
- 3dawarevisionlanguagemodelsfinetuningwithgeometricdistillation
- 3dsmedicaldomainadaptationofllmsviadecomposeddifficultybaseddataselection
- 3mdbenchmedicalmultimodalmultiagentdialoguebenchmark
- 3renhancingsentencerepresentationlearningviaredundantrepresentationreduction
- abenchmarkforhindiverbargumentstructurealternations
- abenchmarkfortranslationsacrossstylesandlanguagevariants
- absvisbenchmarkinghowhumansandvisionlanguagemodelsseeabstractconceptsinimages
- acaseagainstimplicitstandardshomophonenormalizationinmachinetranslationforlanguagesthatusethegeezscript
- acategorytheoreticapproachtoneuralsymbolictaskplanningwithbidirectionalsearch
- acausallensforevaluatingfaithfulnessmetrics
- acceleratedtesttimescalingwithmodelfreespeculativesampling
- accelerateparallelizablereasoningviaparalleldecodingwithinonesequence
- acceleratingllmreasoningviaearlyrejectionwithpartialrewardmodeling
- acceptordenyevaluatingllmfairnessandperformanceinloanapprovalacrosstabletotextserializationapproaches
- accessevalbenchmarkingdisabilitybiasinlargelanguagemodels
- acebenchacomprehensiveevaluationofllmtoolusage
- acingactorcriticforinstructionlearninginblackboxllms
- acloserlookatbiasandchainofthoughtfaithfulnessoflargevisionlanguagemodels
- acomparisonofindependentandjointfinetuningstrategiesforretrievalaugmentedgeneration
- acomprehensiveframeworktooperationalizesocialstereotypesforresponsibleaievaluations
- acomprehensiveliterarychinesereadingcomprehensiondatasetwithanevidencecurationbasedsolution
- acomprehensivesurveyonlearningfromrewardsforlargelanguagemodelsrewardmodelsandlearningstrategies
- acomprehensivesurveyonthetrustworthinessoflargelanguagemodelsinhealthcare
- acomprehensivetaxonomyofnegationfornlpandneuralretrievers
- acomputationalsimulationoflanguageproductioninfirstlanguageacquisition
- acquiescencebiasinlargelanguagemodels
- act2ievaluatingandimprovingactiondepictionintexttoimagemodels
- actionstudioalightweightframeworkfordataandtrainingoflargeactionmodels
- activedomainknowledgeacquisitionwith100dollarbudgetenhancingllmsviacostefficientexpertinvolvedinteractioninsensitivedomains
- activelayercontrastivedecodingreduceshallucinationinlargelanguagemodelgeneration
- activelearningformultidialectalarabicpostagging
- aculturallydiversemultilingualmultimodalvideobenchmarkmodel
- adamsmomentumitselfcanbeanormalizerforllmpretrainingandposttraining
- adaptflowadaptiveworkflowoptimizationviametalearning
- adaptingbiasevaluationtodomaincontextsusinggenerativemodels
- adaptinglargelanguagemodelsforcharacterbasedaugmentativeandalternativecommunication
- adaptivellmroutingunderbudgetconstraints
- adaptivelyprofilingmodelswithtaskelicitation
- adaptiveplattscalingwithcausalinterpretationsforselfreflectivelanguagemodeluncertaintyestimates
- adaptivepreferenceoptimizationwithuncertaintyawareutilityanchor
- adaptiveschemaawareeventextractionwithretrievalaugmentedgeneration
- adaptmergeinferencetimeadaptivevisualandlanguageguidedtokenmergingforefficientlargemultimodalmodels
- adaptthinkreasoningmodelscanlearnwhentothink
- adarewriterunleashingthepowerofpromptingbasedconversationalqueryreformulationviatesttimeadaptation
- adasteeryouralignedllmisinherentlyanadaptivejailbreakdefender
- adatpattentiondebiasedtokenpruningforvideolargelanguagemodels
- additioninfourmovementsmappinglayerwiseinformationtrajectoriesinllms
- addoneinincrementalsampleselectionforlargelanguagemodelsviaachoicebasedgreedyparadigm