Search papers, labs, and topics across Lattice.
6
0
6
5
Evidence-aware retrieval evaluation can change rankings but doesn't guarantee improved answer quality or retriever training, revealing a nuanced relationship between evaluation methods and downstream utility.
Contrastive scoring is key to the success of anchor-based reranking, but the complexity of anchor design may be overhyped, especially with stronger retrievers.
One-third of documents that seem useless to static readers are actually critical for guiding agentic search, revealing a fundamental disconnect in retrieval utility assessments.
Shifting from conceptual to observable relevance can enhance document retrieval effectiveness by up to 10x in non-relevant pruning.
Entity-oriented retrieval's inconsistent performance isn't a model problem, but a data problem: even perfect entity selection only covers 19.7% of relevant documents, and current supervision strategies optimize for semantic relevance at the expense of corpus-grounded discriminativeness.
Escape the scripted feel of simulated conversations: Interplay trains independent user and recommender LLMs that interact in real-time, without pre-defined target items, for more realistic and diverse conversational recommendation data.
