SPLADE-v3: New baselines for SPLADE
read the original abstract
A companion to the release of the latest version of the SPLADE library. We describe changes to the training structure and present our latest series of models -- SPLADE-v3. We compare this new version to BM25, SPLADE++, as well as re-rankers, and showcase its effectiveness via a meta-analysis over more than 40 query sets. SPLADE-v3 further pushes the limit of SPLADE models: it is statistically significantly more effective than both BM25 and SPLADE++, while comparing well to cross-encoder re-rankers. Specifically, it gets more than 40 MRR@10 on the MS MARCO dev set, and improves by 2% the out-of-domain results on the BEIR benchmark.
This paper has not been read by Pith yet.
Forward citations
Cited by 22 Pith papers
-
A Sensitivity-Aware Test Collection for Search Among Personal Information
A new sensitivity-labeled test collection is released from Enron emails with crowdsourced queries, relevance judgments, and LLM extensions for evaluating sensitivity-aware search.
-
HAKARI-Bench: A Lightweight Benchmark for Comparing Retrieval Architectures and Efficiency Settings under Unified Conditions
HAKARI-Bench reconstructs 35 benchmarks into 551 tasks across 43 languages, reproducing full MTEB, MMTEB, and BEIR rankings with Spearman correlation above 0.97 while supporting efficiency variant comparisons.
-
GPUSparse: GPU-Accelerated Learned Sparse Retrieval with Parallel Inverted Indices
GPUSparse delivers exact GPU-accelerated sparse retrieval with 235x CPU speedup on MS MARCO while matching accuracy.
-
Rescaling MLM-Head for Neural Sparse Retrieval
Rescaling the MLM-head projection by a constant factor at initialization resolves scale mismatch in learned sparse retrieval and enables stable use of large-norm backbones such as ModernBERT.
-
Uncertainty-Aware Hybrid Retrieval for Long-Document RAG
UMG-RAG improves long-document RAG by uncertainty-aware fusion of multi-granularity retrievals from complementary dense and sparse retrievers, plus a parent-promotion variant.
-
ColBERTSaR: Sparsified ColBERT Index via Product Quantization
ColBERTSaR uses product quantization on ColBERT embeddings to create a true inverted index that is 50-70% smaller than one-bit PLAID while retaining retrieval effectiveness, and is theoretically equivalent to learned-...
-
Understanding Wacky Weights: A Dissection of SPLADE's Learned Term Importance
SPLADE models produce wacky expansion terms whose prevalence rises with larger vocabularies and falls with stricter sparsity; these terms primarily aid in-domain retrieval rather than out-of-domain generalization.
-
Reproducing Complex Set-Compositional Information Retrieval
Neural retrievers that double BM25 performance on QUEST collapse below 0.02 Recall@100 on the new LIMIT+ benchmark while lexical methods reach 0.96, with all methods degrading as compositional depth increases.
-
Efficient Multivector Retrieval with Token-Aware Clustering and Hierarchical Indexing
TACHIOM speeds up multivector retrieval by up to 247x in clustering and 9.8x in retrieval on MS-MARCOv1 and LoTTE benchmarks using token-distribution-aware centroid allocation and a graph-plus-PQ index, with comparabl...
-
Aligning Dense Retrievers with LLM Utility via Distillation
UAE trains bi-encoder retrievers to match LLM utility distributions via Utility-Modulated InfoNCE, yielding over 30% gains in Recall@1 and MAP on QASPER while running 180x faster than re-ranking.
-
Aligning Dense Retrievers with LLM Utility via Distillation
Distilling LLM perplexity-based utility signals into bi-encoder embeddings via a Utility-Modulated InfoNCE loss yields 30%+ Recall@1 gains over BGE-Base at 180x the speed of LLM re-ranking.
-
Why Advanced Encoders Lag on Sparse Retrieval? The Answer and an Approach to Bridging Vocabulary Gaps
Transferring modern encoders to normalized (lowercased) vocabularies via geometric embedding initialization and activation calibration closes the performance gap in learned sparse retrieval, achieving 52.4 nDCG on BEIR.
-
The LLM Effect on IR Benchmarks: A Meta-Analysis of Effectiveness, Baselines, and Contamination
LLM-enhanced retrieval systems show large effectiveness gains on TREC benchmarks, yet adapted contamination checks indicate some gains may arise from memorization rather than methodological progress.
-
LEAF: Knowledge Distillation of Text Embedding Models with Teacher-Aligned Representations
LEAF distills teacher-aligned student embedding models that achieve new SOTA results on BEIR and MTEB for their size class while requiring only modest data and compute.
-
Improving the Efficiency and Effectiveness of LLM Knowledge Distillation for Conversational Search
Combining contrastive loss with KLD distillation and adding sparsity regularization improves effectiveness and reduces FLOPS by 2x in conversational search with minimal recall loss.
-
Surface-Form Neural Sparse Retrieval: Robust Fuzzy Matching for Industrial Music Search
A neural sparse retrieval system with granular subword tokenization (max 3 chars) achieves 91.4% recall@10 on a 6M music document corpus versus 57.7% for trigrams, with improved HCI exploration efficiency and zero add...
-
Efficient Listwise Reranking with Compressed Document Representations
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.
-
BioHiCL: Hierarchical Multi-Label Contrastive Learning for Biomedical Retrieval with MeSH Labels
BioHiCL applies hierarchical multi-label contrastive learning with MeSH annotations to improve biomedical retrieval, sentence similarity, and question answering using small efficient models.
-
Empirical Evaluation of PDF Parsing and Chunking for Financial Question Answering with RAG
Systematic tests show that specific PDF parsers combined with overlapping chunking strategies better preserve structure and improve RAG answer correctness on financial QA benchmarks including the new TableQuest dataset.
-
Beyond Relevance: On the Relationship Between Retrieval and RAG Information Coverage
Coverage-focused retrieval metrics correlate strongly with nugget coverage in RAG responses across text and multimodal benchmarks, supporting their use as performance proxies when retrieval and generation goals align.
-
Overview of the TalentCLEF 2026: Skill and Job Title Intelligence for Human Capital Management
The paper describes the organization, tasks, datasets, and participation results for the TalentCLEF 2026 challenge, which received 113 team registrations and over 400 submissions.
-
A Survey on Retrieval-Augmented Text Generation for Large Language Models
A survey that categorizes RAG methods for LLMs into four retrieval-centric stages, reviews their evolution and evaluation, and outlines challenges and future directions.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.