Dense retrievers favor passages naming high-scoring answer objects
Across eight models, higher token-prediction scores predict higher retrieval ranks even when the relation is wrong.
Stories per day · 30 days
2 RSS
Relatedembeddings2contrastive-learning1dense-retrieval1evidence-synthesis1evolutionary-search1generative-retrieval1hard-negatives1instruction-following1language-models1msmarco1open-weight1qwen31search-bias1semantic-identifiers1systematic-review1
11 stories, newest first · one card per story
Stories tagged #information-retrieval
Embedding search finds 92% of trial evidence versus 45% for labels
On 7,354 randomized trials, a two-stage embedding-plus-language-model filter recovers far more systematic-review evidence, with cost rising by search depth.
NHNM mines natural hard negatives for instruction retrieval
Fine-tuning 568M to 4B models on nearly 7K NHNM-MARCO instances matches or beats much larger retrievers on InstructIR and FollowIR.
LLM evolutionary search improves generative retrieval semantic indexes
On NQ320K, jointly evolving SID construction and collision handling lifts Recall@1 from 55.8% to 59.3%.
Can Language Models Play the Wikipedia Game?
GPT-4o, o3-mini, QwQ-32B, and DeepSeek-R1-70B matched humans at Wikipedia-link navigation without clearly beating them.
A Systematic Study of Semantic ID Spaces for Generative Information Retrieval
Study unifies product and residual quantization for semantic DocIDs and tests them on MS MARCO and NQ.
Effective Dense Retrieval using Only In-Context Examples
RICE turns decoder-only LLMs into training-free dense retrievers by conditioning embeddings on shared in-context examples.
Retrieving Biblical Intertextual References in Karen Blixen's Seven Gothic Tales
Fine-tuned DFM-large retrieves biblical allusions in Blixen at R@10 0.508 on a 189-reference benchmark.
ANTMAN: Adaptive Need Tracking for Multi-Agent Navigation in Large Information Spaces
ANTMAN's Need Graph keeps multi-agent coordination nearly flat, rising 1.23x when searchable context grows 16x.
A Living Benchmark for Information Retrieval from Electronic Health Records
Clinicians validate BRIE, an auto-generated EHR benchmark showing leading LLMs often omit clinically important facts.
Think Thrice Before Reranking: Multi-perspective Evidence and Reasoning Integration for Text Reranking
MERIT-Rank integrates multiple LLM reasoning trajectories for text reranking; its 4B model outperforms most 7B and 32B rerankers on BRIGHT.