Pith. sign in

REVIEW 13 cited by

Fine-Tuning LLaMA for Multi-Stage Text Retrieval

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2310.08319 v1 pith:KJ4KQ32N submitted 2023-10-12 cs.IR

Fine-Tuning LLaMA for Multi-Stage Text Retrieval

classification cs.IR
keywords modelsretrievaleffectivenesslanguagellmsdemonstratefine-tuninglarge
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

The effectiveness of multi-stage text retrieval has been solidly demonstrated since before the era of pre-trained language models. However, most existing studies utilize models that predate recent advances in large language models (LLMs). This study seeks to explore potential improvements that state-of-the-art LLMs can bring. We conduct a comprehensive study, fine-tuning the latest LLaMA model both as a dense retriever (RepLLaMA) and as a pointwise reranker (RankLLaMA) for both passage retrieval and document retrieval using the MS MARCO datasets. Our findings demonstrate that the effectiveness of large language models indeed surpasses that of smaller models. Additionally, since LLMs can inherently handle longer contexts, they can represent entire documents holistically, obviating the need for traditional segmenting and pooling strategies. Furthermore, evaluations on BEIR demonstrate that our RepLLaMA-RankLLaMA pipeline exhibits strong zero-shot effectiveness. Model checkpoints from this study are available on HuggingFace.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 13 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. The Voronoi Bottleneck: Capacity-Aware Dense Retrieval for Product Search

    cs.IR 2026-06 unverdicted novelty 7.0

    Proves Voronoi complexity equals sign-rank for top-1 retrieval, introduces CUS diagnostic predicting retrieval failure at AUC >0.8 without labels, and AT-DW-InfoNCE objective with derived alpha^*=2.0 that improves Rec...

  2. Re-Ranking Through an Attribution Lens for Citation Quality in Legal QA

    cs.CL 2026-06 unverdicted novelty 7.0

    Re-ranking retrieval candidates via a cross-encoder trained on continuous perturbation-based attribution scores improves citation faithfulness and gold-answer alignment in legal QA over semantic similarity.

  3. Test-Time Compute for Frozen Embedding Models through Agentic Program Search

    cs.LG 2026-05 unverdicted novelty 7.0

    Agentic program search over frozen embedding APIs yields a parameter-free inference algebra—a softmax-weighted centroid of top-K documents interpolated with the query—that lifts nDCG@10 across seven model families on ...

  4. Test-Time Compute for Frozen Embedding Models through Agentic Program Search

    cs.LG 2026-05 unverdicted novelty 7.0

    A softmax-weighted centroid of the local top-K documents interpolated with the query improves nDCG@10 for frozen embedding models across seven families on held-out BEIR data.

  5. Test-Time Compute for Frozen Embedding Models through Agentic Program Search

    cs.LG 2026-05 unverdicted novelty 6.0

    Agentic program search over a frozen encoder API yields retrieval programs that improve nDCG@10 on held-out tasks and unseen encoder families with no per-domain training.

  6. Linguistic Nepotism: Trading-off Quality for Language Preference in Multilingual RAG

    cs.CL 2025-09 conditional novelty 6.0

    In multilingual retrieval-augmented generation, models cite English evidence more accurately than translated evidence, and this language preference can outweigh document relevance.

  7. RankZephyr: Effective and Robust Zero-Shot Listwise Reranking is a Breeze!

    cs.IR 2023-12 conditional novelty 6.0

    RankZephyr is a new open-source LLM that closes the effectiveness gap with GPT-4 for zero-shot listwise reranking while showing robustness to input ordering and document count.

  8. One Retrieval to Cover Them All: Co-occurrence-Aware Knowledge Base Reorganization for Session-Level RAG

    cs.IR 2026-06 unverdicted novelty 5.0

    Co-occurrence-aware KB clustering raises session-level RAG coverage from 41% to 58% on WixQA while cutting retrieval calls and compressing the KB to 20% of original size.

  9. Efficient Listwise Reranking with Compressed Document Representations

    cs.IR 2026-04 unverdicted novelty 5.0

    RRK compresses documents to multi-token embeddings for efficient listwise reranking, enabling an 8B model to achieve 3x-18x speedups over smaller models with comparable or better effectiveness.

  10. Dynamic Ranked List Truncation for Reranking Pipelines via LLM-generated Reference-Documents

    cs.IR 2026-04 unverdicted novelty 5.0

    LLM-generated reference documents serve as relevance pivots for dynamic ranked-list truncation and adaptive/parallel listwise reranking, reportedly beating prior RLT methods and cutting LLM reranking cost by up to 66%.

  11. Dynamic Ranked List Truncation for Reranking Pipelines via LLM-generated Reference-Documents

    cs.IR 2026-04 unverdicted novelty 5.0

    LLM-generated reference documents enable dynamic ranked list truncation and adaptive batching for listwise reranking, outperforming prior RLT methods and accelerating processing by up to 66% on TREC benchmarks.

  12. Beyond Hard Negatives: The Importance of Score Distribution in Knowledge Distillation for Dense Retrieval

    cs.IR 2026-04 unverdicted novelty 5.0

    Stratified sampling preserving teacher score distribution outperforms hard-negative mining as a robust baseline for knowledge distillation in dense retrieval.

  13. RAG-Match: Retrieval-Augmented Knowledge Injection and Hierarchical Reasoning for Calibrated Semantic Relevance

    cs.IR 2026-05 unverdicted novelty 4.0

    RAG-Match is a three-stage framework for semantic relevance modeling that integrates knowledge-augmented pretraining, hierarchical reasoning alignment, and preference-based decision calibration, outperforming LLM base...