There are 0 repository under cross-encoder topic.
Lite & Super-fast re-ranking for your search & retrieval pipelines. Supports SoTA Listwise and Pairwise reranking based on LLMs and cross-encoders and more. Created by Prithivi Da, open for PRs & Collaborations.
Testing speed and accuracy of RAG with, and without Cross Encoder Reranker.
Sentence-Transformers Information Retrieval example on Chinese
Dual Cross Encoder for Dense Retrieval
An overview of popular reranking models and architectures for 2 stage RAG pipelines
Livebook to run a Phoenix_LiveView documentation Retrieval-Augmented Generation (RAG) enhanced LLM
Improving Bilingual Lexicon Induction with Cross-Encoder Reranking (Findings of EMNLP 2022). Keywords: Bilingual Lexicon Induction, Word Translation, Cross-Lingual Word Embeddings.
Semantically Search Emojis From the Command Line!
Semantic Emoji Search Plugin for FiftyOne
A genral RAG Search chatbot, with SoTA RAG techniques such as HyDE, Hybrid retrieval with BM25 + RRF and Cross encoder reranking. Evaluated on the BEIR scifact dataset and compared all the different pipelines i tried along the way
This repository hosts the code to launch a streamlit Q&A app that locally uses LLMs in a RAG-Reranker workflow
A hybrid Research Assistant that combines an exact Knowledge Graph (Neo4j) with a Retrieval‑Augmented Generation pipeline (FAISS + Cross‑Encoder + FLAN‑T5) behind a sleek Streamlit interface.
This repository showcases a comprehensive approach to information retrieval, document re-ranking, and language model integration. It incorporates techniques such as document chunking, embedding projection, and automatic query expansion to enhance the effectiveness of information retrieval systems.
Leveraging torch.compile to accelerate cross-encoder inference
PastProof AI – ML core for automated fact-checking: ingests raw text, finds evidence in a custom corpus, and returns only the false claims it can refute, together with supporting passages and (optionally) an LLM explanation.
A GenAI based search system that scans numerous fashion product descriptions to recommend suitable options based on user queries.
An AI analyst with a hybrid LLM architecture 🤖. Uses a fine-tuned Phi-3 Mini (3.8B) for local RAG answer generation & Gemini 1.5 Pro for query analysis of SEC filings (AAPL, MSFT, GOOG, AMZN, META).
TinyBERT-based bi-encoder, cross-encoder, and poly-encoder trained on MS MACRO for passage re-ranking
End-to-end PDF RAG: FastAPI + Streamlit UI, Qdrant, and RAG workflows powered by LangChain/LangGraph. Dockerized with caching, optional GPU, and Prometheus/Grafana/Loki.
Official respository for Legal Tutorial in The 14th European Summer School on Information Retrieval
AI-powered document originality and plagiarism risk detection system combining semantic similarity (SBERT), stylometric analysis, and citation masking for explainable, multi-layered originality scoring.
Collection of some of my works during my internship period at Salahkart for preview and educational purpose only.
Baseline models for searching for movie plots from Wikipedia articles. Techniques include BM25 (lexical search), bi/cross-encoding (semantic search), and retrieval-augmented generation (RAG) using Mistal 7B through Fireworks.ai.
Retrieval-Augmented QA system for research documents/insurance policies using LangChain, ChromaDB, and OpenAI LLMs. Supports query reranking and few-shot prompting.
Timeline Project
A RAG agent for local document retrieval and answer generation
AniSearchModel leverages Sentence-BERT (SBERT) models to generate embeddings for synopses, enabling the calculation of semantic similarities between descriptions. This allows users to find the most similar anime or manga based on a given description.
A Web App for Searching Steam Games
Prototipe Document Q&A Bot yang dirancang untuk menyajikan informasi terkait regulasi.
Ask questions to your PDFs using AI (RAG + Streamlit). Upload any PDF and get smart answers — powered by semantic search and summarization.
LLM News & Research Radar : Self-Improving RAG Engine
Évaluation de la pertinence (question ↔ article juridique) en français. Pipeline complet (prépa → modèles → soumission) avec CamemBERT en bi-encodeur calibré (MSE/Spearman), + variantes cross-encoder.
Hybrid RAG for Thai ROV patch notes: normalize diffs → hybrid retrieve (BM25 + dense in Chroma, MMR, optional bge reranker) → citation-first answers via FastAPI with OpenAI/Ollama generators. | ระบบ RAG ภาษาไทยสำหรับแพตช์โน้ต ROV
Set of notebooks used to experiment and learn about retreival and reranking strategies provided by qdrant over BEIR MS-Macros dataset.
🔍 Analyze system prompts in large language models to understand design principles and enhance AI application effectiveness.
Exploration of retrieval methods on the HotpotQA corpus, combining dense retrieval and feature-based reranking. Achieved a mean nDCG@10 of 0.9416 using LambdaRank with features such as cross-encoder score, LLM score, BM25 score, and token-based statistics—surpassing dense retriever + cross-encoder baselines.