LambdaRank
PulseAugur coverage of LambdaRank — every cluster mentioning LambdaRank across labs, papers, and developer communities, ranked by signal.
-
Pairwise ranking beats RL for LLM explanation selection in recommendation systems
Researchers have developed a new method for selecting explanations from large language models (LLMs) in recommendation systems, significantly reducing serving costs and latency. By pre-generating a pool of explanations …
-
New study predicts Bluesky custom feed returns for posts
Researchers have developed a new method for predicting which custom feeds on the Bluesky platform will surface new posts. This approach treats new posts as queries and ranks existing feeds based on their likelihood of r…
-
New RISC method boosts LLM accuracy by ranking answers
Researchers have developed a new method called Ranking-Improved Self-Consistency (RISC) to enhance the accuracy of large language models. This approach reframes the selection of answers from multiple generated reasoning…
-
AI explanation quality assessed via ranking, outperforming regression
Researchers have reframed the evaluation of AI explanation quality from a generation task to a ranking problem. Instead of producing a single best explanation, models are trained to discern the relative quality among mu…