Pith. sign in

REVIEW 25 cited by

Generative Representational Instruction Tuning

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 2402.09906 v3 pith:PEHKX6B3 submitted 2024-02-15 cs.CL cs.AIcs.LG

Generative Representational Instruction Tuning

classification cs.CL cs.AIcs.LG
keywords modelsgenerativeembeddinggenerationgritgritlmlanguageother
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

All text-based language problems can be reduced to either generation or embedding. Current models only perform well at one or the other. We introduce generative representational instruction tuning (GRIT) whereby a large language model is trained to handle both generative and embedding tasks by distinguishing between them through instructions. Compared to other open models, our resulting GritLM 7B sets a new state of the art on the Massive Text Embedding Benchmark (MTEB) and outperforms all models up to its size on a range of generative tasks. By scaling up further, GritLM 8x7B outperforms all open generative language models that we tried while still being among the best embedding models. Notably, we find that GRIT matches training on only generative or embedding data, thus we can unify both at no performance loss. Among other benefits, the unification via GRIT speeds up Retrieval-Augmented Generation (RAG) by > 60% for long documents, by no longer requiring separate retrieval and generation models. Models, code, etc. are freely available at https://github.com/ContextualAI/gritlm.

discussion (0)

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

Forward citations

Cited by 25 Pith papers

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

  1. IdioLink: Retrieving Meaning Beyond Words Across Idiomatic and Literal Expressions

    cs.CL 2026-05 unverdicted novelty 7.0

    IdioLink introduces a benchmark dataset and evaluation showing that strong embedding models struggle to retrieve equivalent meanings across idiomatic and literal forms, relying on shallow cues instead.

  2. 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 ...

  3. 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.

  4. Rethinking Reasoning-Intensive Retrieval: Evaluating and Advancing Retrievers in Agentic Search Systems

    cs.CL 2026-05 unverdicted novelty 7.0

    BRIGHT-Pro and RTriever-Synth advance reasoning-intensive retrieval by adding multi-aspect evidence evaluation and aspect-decomposed synthetic training, with the fine-tuned RTriever-4B showing gains over its base model.

  5. Autonomous Knowledge Graph Exploration with Adaptive Breadth-Depth Retrieval

    cs.AI 2026-01 unverdicted novelty 7.0

    ARK adaptively retrieves from knowledge graphs using global lexical search and one-hop neighborhood exploration, reaching 59.1% Hit@1 on STaRK with up to 31.4% gains over training-free baselines and enabling distillat...

  6. UEmbed: Unified Sparse and Dense Multimodal Embeddings

    cs.CV 2026-08 conditional novelty 6.0

    UEmbed uses 16 special tokens over a partitioned vocabulary to make a decoder-only multimodal model emit dense and sparse embeddings in one forward pass; the 9B model scores 71.8 dense / 71.0 sparse on MMEB-v2.

  7. FiRE: Enhancing MLLMs with Fine-Grained Context Learning for Complex Image Retrieval

    cs.CV 2026-07 conditional novelty 6.0

    Disentangled fine-grained context then retrieval fine-tuning on an 87K auto-generated quintuple CIR dataset lifts a 4B MLLM past larger universal retrievers on complex zero-shot image search.

  8. Semantic Retrieval for Product Search in E-Commerce

    cs.IR 2026-05 unverdicted novelty 6.0

    A dual-encoder LLM is trained via contrastive learning then ROAR to retrieve exact matches and rank substitutes in e-commerce search.

  9. Layer-wise Representation Dynamics: An Empirical Investigation Across Embedders and Base LLMs

    cs.LG 2026-05 unverdicted novelty 6.0

    LRD framework with Frenet, NRS, and GFMI metrics shows layer-wise structure in 31 models provides usable signal for model selection and pruning on MTEB tasks.

  10. 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.

  11. Reproducing Complex Set-Compositional Information Retrieval

    cs.CL 2026-05 unverdicted novelty 6.0

    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.

  12. Efficient Rationale-based Retrieval: On-policy Distillation from Generative Rerankers based on JEPA

    cs.IR 2026-04 unverdicted novelty 6.0

    Rabtriever distills a generative reranker into an efficient bi-encoder using on-policy JEPA to achieve near-reranker accuracy with linear complexity on rationale-based retrieval.

  13. Efficient Rationale-based Retrieval: On-policy Distillation from Generative Rerankers based on JEPA

    cs.IR 2026-04 unverdicted novelty 6.0

    Rabtriever distills a generative reranker into an efficient independent encoder using JEPA and auxiliary reverse KL loss to achieve linear complexity and strong performance on rationale-based retrieval tasks.

  14. Explainable Disentangled Representation Learning for Generalizable Authorship Attribution in the Era of Generative AI

    cs.CL 2026-04 unverdicted novelty 6.0

    EAVAE disentangles style from content via contrastive pretraining and an explainable discriminator in a VAE setup, claiming SOTA authorship attribution on multiple datasets and strong few-shot AI text detection.

  15. Modular Representation Compression: Adapting LLMs for Efficient and Effective Recommendations

    cs.IR 2026-04 unverdicted novelty 6.0

    LLMs exhibit mid-layer representation advantage for recommendations; MARC compresses representations modularly to reduce costs while improving performance, as shown in a large-scale online advertising deployment.

  16. ViLL-E: Video LLM Embeddings for Retrieval

    cs.CV 2026-04 unverdicted novelty 6.0

    ViLL-E introduces a dynamic embedding mechanism and joint contrastive-generative training for VideoLLMs, delivering up to 7% gains in temporal localization and 4% in video retrieval while enabling new zero-shot capabilities.

  17. OPERA: Online Data Pruning for Efficient Retrieval Model Adaptation

    cs.IR 2026-03 conditional novelty 6.0

    Dynamic hierarchical data pruning improves NDCG@10 and Recall@20 for dense retrievers while reaching full performance in half the iterations.

  18. EmbeddingGemma: Powerful and Lightweight Text Representations

    cs.CL 2025-09 unverdicted novelty 6.0

    A 300M-parameter open embedding model sets new SOTA on MTEB for its size class and matches models twice as large while staying effective when compressed.

  19. Should We Still Pretrain Encoders with Masked Language Modeling?

    cs.CL 2025-07 accept novelty 6.0

    Controlled ablations of 38 models find MLM superior to CLM on representation benchmarks while CLM offers better data efficiency and stability; a biphasic CLM-then-MLM schedule is optimal under fixed compute and improv...

  20. DataComp-LM: In search of the next generation of training sets for language models

    cs.LG 2024-06 unverdicted novelty 6.0

    DCLM-Baseline dataset lets a 7B model reach 64% 5-shot MMLU accuracy after 2.6T tokens, beating prior open-data models by 6.6 points on MMLU with 40% less compute.

  21. NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models

    cs.CL 2024-05 accept novelty 6.0

    NV-Embed achieves first place on the MTEB leaderboard across 56 tasks by combining a latent attention layer, causal-mask removal, two-stage contrastive training, and data curation for LLM-based embedding models.

  22. StarCoder 2 and The Stack v2: The Next Generation

    cs.SE 2024-02 accept novelty 6.0

    StarCoder2-15B matches or beats CodeLlama-34B on code tasks despite being smaller, and StarCoder2-3B outperforms prior 15B models, with open weights and exact training data identifiers released.

  23. SHIFT: Self-reconstruction Harnesses Implicit Fine-grained Thinking for Retrieval

    cs.IR 2026-07 conditional novelty 5.0

    SHIFT trains LLM retrievers to reason through latent continuous tokens and reconstruct explicit reasoning traces, improving reasoning-intensive retrieval.

  24. Robustness Risk of Conversational Retrieval: Identifying and Mitigating Noise Sensitivity in Qwen3-Embedding Model

    cs.IR 2026-02 unverdicted novelty 5.0

    Qwen3-embedding models show noise sensitivity in conversational retrieval where dialogue artifacts rank highly despite lacking semantic value, a problem reduced by query prompting and more severe than in prior Qwen ve...

  25. MLT-Dedup: Efficient Large-Scale Online Video Deduplication via Multi-Level Representations and Spatial-Temporal Matching

    cs.CV 2026-06 unverdicted novelty 4.0

    MLT-Dedup achieves 91% reduction in online video repetition rates at 90% precision and 5x indexing capacity using multi-level representations and differential feature-enhanced similarity on a real-world platform.