Spaces:
Running
Running
metadata
title: README
emoji: 馃悪
colorFrom: blue
colorTo: indigo
sdk: static
pinned: true
SightAndCite
Retrieval that sees the structure and cites the evidence.
Open research on retrieval-augmented generation (RAG), chunking, multi-hop reasoning and efficient LLM systems.
馃敪 Featured: SpectRAG
Stop retrieving chunks. Start retrieving context. SpectRAG replaces independent top-k ranking with context-set selection over a weighted chunk graph and its Laplacian spectrum. It returns evidence that is relevant, diverse and connected, all under a fixed token budget.
| 馃搫 Paper | arXiv:2610.36991 路 馃 Paper page |
| 馃幀 Visual walkthrough | sightandcite/SpectRAG |
| 馃搳 Evaluation results | sightandcite/SpectRAG-eval-results |
| 馃捇 Code (MIT) | github.com/SightAndCite/SpectRAG |
Highlights on MultiHopRAG & UltraDomain
- 91.8% overall gold-evidence recall (next best 76.1%)
- 95.1% exact match on inference questions
- ~6脳 smaller index than LightRAG (46 MB vs 274 MB)
- LLM judges prefer SpectRAG's answers for diversity against every baseline
馃Л What we work on
- Retrieval-augmented generation: context-set selection, hybrid retrieval, graph-based RAG
- Chunking: structure-aware and semantic segmentation for long documents
- Multi-hop systems: recovering bridge evidence across chunks and documents
- Efficient LLM pipelines: distillation, lightweight indexes, fully local deployments
馃懃 Team
Zafer Sahinoglu 路 Abdulazeez Mohammed 路 Buse Yaren Kazangirler 路 Cahit Berkay Kazangirler
馃摤 Get in touch
Questions, ideas or a retriever that breaks in interesting ways? Open a discussion on the paper page or an issue on GitHub.