apple ml research · 23 topics

glyph multi strategy agentic system

shared selective persistent memory for agentic llm systems

discosign discourse aware text to sign language gloss translation

agent seer synthesizing scenarios

from preferences to principles

grpo beyond english

when unlearning is free leveraging low influence points to reduce computational costs

scaling categorical flow maps

apple taming outlier tokens in diffusion transformers

lead breaking the no recovery bottleneck

show me examples inferring visual concepts from image sets

when unlearning is free

one layer is enough adapting pretrained visual encoders

clara bridging retrieval and generation

multilingual semantic retrieval for apple music search

behavioral privacy leakage in agentic negotiation

incentivizing temporal awareness in egocentric video understanding models

anti causal domain generalization

dynamics fine tuning llms

lensvlm selective context expansion

revisiting asr error correction with specialized models

fortress a case study in stabilizing search recommendations

path constrained mixture of experts