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

arXiv: https://arxiv.org/abs/2607.21333 · HTML: https://mlsanity.com/paper/2607.21333/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Knowledge- and tool-augmented reasoning, one of 17 directions in LLM reasoning.
- Builds on 20 of its 44 references that are in the map.
- First posted on arXiv 23 Jul 2026.

## Placed in

- LLM reasoning / [Knowledge- and tool-augmented reasoning](https://mlsanity.com/lr/knowledge-and-tool-augmented-reasoning/)

## Builds on (most cited first, 20 of 20)

- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models](https://mlsanity.com/paper/2506.05176/) (2025)
- [Text Embeddings by Weakly-Supervised Contrastive Pre-training](https://mlsanity.com/paper/2212.03533/) (2022)
- [M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation](https://mlsanity.com/paper/2402.03216/) (2024)
- [Training Large Language Models to Reason in a Continuous Latent Space](https://mlsanity.com/paper/2412.06769/) (2024)
- [BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent](https://mlsanity.com/paper/2508.06600/) (2025)
- [Implicit Chain of Thought Reasoning via Knowledge Distillation](https://mlsanity.com/paper/2311.01460/) (2023)
- [Think Then Embed: Generative Context Improves Multimodal Embedding](https://mlsanity.com/paper/2510.05014/) (2025)
- [Expediting and Elevating Large Language Model Reasoning via Hidden Chain-of-Thought Decoding](https://mlsanity.com/paper/2409.08561/) (2024)
- [ReasonEmbed: Enhanced Text Embeddings for Reasoning-Intensive Document Retrieval](https://mlsanity.com/paper/2510.08252/) (2025)
- [TongSearch-QR: Reinforced Query Reasoning for Retrieval](https://mlsanity.com/paper/2506.11603/) (2025)
- [Let LLMs Speak Embedding Languages: Generative Text Embeddings via Iterative Contrastive Refinement](https://mlsanity.com/paper/2509.24291/) (2025)
- [LaSER: Internalizing Explicit Reasoning into Latent Space for Dense Retrieval](https://mlsanity.com/paper/2603.01425/) (2026)
- [GRACE: Generative Representation Learning via Contrastive Policy Optimization](https://mlsanity.com/paper/2510.04506/) (2025)
- [Your Dense Retriever is Secretly an Expeditious Reasoner](https://mlsanity.com/paper/2510.21727/) (2025)
- [Search-R3: Unifying Reasoning and Embedding in Large Language Models](https://mlsanity.com/paper/2510.07048/) (2025)
- [Compressing then Matching: An Efficient Pre-training Paradigm for Multimodal Embedding](https://mlsanity.com/paper/2511.08480/) (2025)
- [Pretraining with Token-Level Adaptive Latent Chain-of-Thought](https://mlsanity.com/paper/2602.08220/) (2026)
- [CREM: Compression-Driven Representation Enhancement for Multimodal Retrieval and Comprehension](https://mlsanity.com/paper/2602.19091/) (2026)
- [Reconstructing Content with Collaborative Attention for Universal Multimodal Representation Learning](https://mlsanity.com/paper/2603.01471/) (2026)

## Abstract (arXiv)

LLM-based retrievers have become a fundamental component of modern information retrieval systems. The paradigm of "rewrite-then-retriev" introduces explicit reasoning before retrieval. In addition, implicit-reasoning retrievers such as GIRCSE and LaSER improve efficiency by replacing explicit reasoning with soft tokens. Although these methods demonstrated competitive performance on reasoning-intensive retrieval benchmarks, they struggle to address the mismatch between the objectives of retrieval and generation. In this work, we propose SHIFT (Self-reconstruction Harnesses Implicit Fine-grained Thinking for Retrieval), a retrieval training framework based on LLMs. Firstly, we transfer LLMs into reasoning-efficient retrievers with residual projection and task-oriented bidirectional attention aggregation in the latent space. Secondly, we alleviate the mismatch between contrastive learning and implicit reasoning using fine-grained next-token-prediction-based reconstruction. Extensive experiments on reasoning-intensive retrieval benchmarks show that SHIFT consistently outperforms other widely used retrievers. We also carried out a detailed analysis to illustrate how our method works.