Fine-tuning a LLaMA model with GRPO on live search-engine feedback yields query expansions that retrieve more relevant and diverse product results in e-commerce search.
On a notion of oplax 3-functor
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
We introduce a notion of normalised oplax $3$-functor suitable for the elementary homotopy theory of strict $3$-categories, following the combinatorics of orientals. We show that any such morphism induces a morphism of simplicial sets between the Street nerves and we characterise those morphisms of simplicial sets coming from normalised oplax $3$-functors. This allows us to prove that normalised oplax $3$-functors compose. Finally we construct a strictification for normalised oplax $3$-functors whose source is a $1$-category without split-monos or split-epis.
citation-role summary
citation-polarity summary
fields
cs.IR 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
LESER: Learning to Expand via Search Engine-feedback Reinforcement in e-Commerce
Fine-tuning a LLaMA model with GRPO on live search-engine feedback yields query expansions that retrieve more relevant and diverse product results in e-commerce search.