2026-11-02 ~ 11-08 · AI Search
Generative · Personalization · AI Search · Infra · Embodied AI
- 1. Contextual Embeddings Beyond the Gold Chunk · 📖 Deep dive
When RAG splits a long document into chunks, answering a question often needs more than the chunk that states the answer: it also needs the passages saying which product or which year the answer is about…
Appeared in: 11-02
- 2. DeFT-RAG: Incrementally Decomposing Multi-Hop Questions into Self-Contained Sub-Questions with Reusable Answers · 📖 Deep dive
Multi-hop questions need evidence found link by link through intermediate entities that the question never names. Graph-based RAG starts from the entities the question mentions and mostly reaches only the first link…
Appeared in: 11-02
- 3. Decompose, Estimate, Select: Coverage-Aware Context Construction for Long-Form RAG
Long-form RAG needs a set of documents that complement each other, but the ten documents picked by scoring relevance one by one often cover the same aspect again and again and leave others uncovered…
Appeared in: 11-02
- 4. Glance: The Generator Is Its Own Reranker
RAG systems usually serve a reranker and a generator side by side, and the candidate passages are read once by each model. Glance starts from the idea that the generator reads these passages anyway…
Appeared in: 11-02
- 5. Retrieving from Your Own Mistakes: On-Policy Training of Embedding Models for Multi-Hop Retrieval-Augmented Generation · 📖 Deep dive
A chain embedding retriever for multi-hop retrieval sees only correct passages in its retrieval history during training, but in use it sees the passages it found itself, which are often wrong…
Appeared in: 11-02
- 6. xPRT: Cross-Passage Retrieval Pretraining
In an unlabeled web page, which passages talk about related things is itself usable for training a retriever. xPRT works like this: it first picks passages whose vectors are closest to the average vector of the whole…
Appeared in: 11-02
- 7. How Specific Do I Need to Be? Effects of User Query Underspecification on Deep Research Agents
When users ask a deep research agent a question, they often leave out the time range, use the wrong technical term, or skip their preferences. The authors remove or blur each kind of information in fully specified…
Appeared in: 11-02
- 8. Quantum-Inspired Pairwise Ranking Aggregation via Active LLM Feedback
When fusing results from several retrievers, unsupervised methods such as RRF are cheap but do not learn, while listwise LLM reranking works well but needs hundreds of calls per query. LLM-IRA takes a middle path…
Appeared in: 11-02
- 9. Relevance-Guided Residual Calibration for Spectral Compression of Dense Retrieval Embeddings
Projecting retrieval embeddings onto a few principal directions is a common way to cut storage and scoring cost, but how much weight each direction keeps follows only an overall trend from corpus statistics…
Appeared in: 11-02
- 10. Feedback-Assisted Trust Propagation over Document Relation Graphs for Retrieval-Augmented Generation
RAG corpora often contain outdated, contradictory or unreliable documents, and labeling which ones are trustworthy by hand is expensive, so labels cover only a small part. This paper treats each document as a node in a…
Appeared in: 11-02