Shuyue Stella Li
PulseAugur coverage of Shuyue Stella Li — every cluster mentioning Shuyue Stella Li across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
New HorizonBench benchmark tests AI's ability to track evolving user preferences
Researchers have introduced HorizonBench, a new benchmark designed to evaluate how well AI personal assistants can adapt to users' evolving preferences over long periods. Unlike previous benchmarks, HorizonBench simulat…
-
New CAPE method enhances LLM personalization with Bayesian inference
Researchers have developed CAPE (Cold-start Adaptive Preference Elicitation), a novel method for personalizing large language model (LLM) systems when minimal user data is available. CAPE leverages structured world mode…
-
New research benchmarks LLM reasoning, safety, and mathematical failures
Recent research explores the reasoning capabilities and limitations of large language models (LLMs). One paper introduces a framework to analyze LLM reasoning by synthesizing cognitive science research, identifying diff…
-
New benchmarks and methods tackle video reward modeling without human feedback · 2 sources tracked
Researchers have developed new methods for training video reward models without relying on human feedback. IsoVRBench introduces a benchmark of synthetic preference pairs that isolate specific video properties, leading …
-
ExploreNet enhances diffusion models by learning adaptive noise scales
Researchers have developed ExploreNet, a novel policy designed to improve image generation by learning an adaptive exploration distribution within diffusion models. Unlike previous methods that applied uniform noise, Ex…