Diffusion vs. Autoregressive Language Models: A Text Embedding Perspective (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language model (LLM)-based embedding models surpass BERT and T5 on general-purpose text embeddable tasks. |
| Approach: | They propose to adopt diffusion language models for text embeddings to overcome limitations in unidirectional attention used during autoregressive pre-training. |
| Outcome: | The proposed model outperforms the existing LLM-based embedding model on reasoning tasks by 20% and 2% on traditional embeddable benchmarks. |
Similar Papers
Learning to Look at the Other Side: A Semantic Probing Study of Word Embeddings in LLMs with Enabled Bidirectional Attention (2025.acl-long)
Copied to clipboard
| Challenge: | Autoregressive Large Language Models (LLMs) demonstrate exceptional performance in language understanding and generation tasks, but their application in text embedding tasks has been relatively slow due to the constraints of the unidirectional attention mechanism. |
| Approach: | They propose to enable bidirectional attention in LLMs to improve their ability to represent subsequent context but weaken their utilization of preceding context. |
| Outcome: | The proposed model improves the LLMs’ ability to represent subsequent context but weakens their utilization of preceding context, while contrastive learning training can help to maintain both abilities. |
Enhancing Lexicon-Based Text Embeddings with Large Language Models (2025.acl-long)
Copied to clipboard
| Challenge: | Recent large language models (LLMs) have demonstrated exceptional performance on general-purpose text embedding tasks. |
| Approach: | They introduce the first lexicon-based embeddings that consolidates the vocabulary space through token embeddation clustering to handle the issue of token redundancy in LLM vocabularies. |
| Outcome: | The proposed model outperforms dense embeddings on the Massive Text Embedding Benchmark (MTEB) it also supports efficient dimension pruning without any specialized objectives like Matryoshka Representation Learning. |
Llama2Vec: Unsupervised Adaptation of Large Language Models for Dense Retrieval (2024.acl-long)
Copied to clipboard
| Challenge: | Dense retrieval requires discriminative embeddings to represent the semantic relationship between query and document. |
| Approach: | They propose an unsupervised approach that performs unsupervised adaptation of large language models for dense retrieval. |
| Outcome: | The proposed model improves on a variety of dense retrieval benchmarks and is available on github. |
BeLLM: Backward Dependency Enhanced Large Language Model for Sentence Embeddings (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing LLMs adopt autoregressive architectures without explicit backward dependency modeling. |
| Approach: | They propose a backward dependency enhanced large language model that transforms attention layers from uni-to-bi-directional to learn sentence embeddings. |
| Outcome: | The proposed model achieves state-of-the-art performance in varying scenarios. |
Co-Evolving LLMs and Embedding Models via Density-Guided Preference Optimization for Text Clustering (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for text clustering use static pseudo-oracles, i.e., unidirectionally querying them for similarity assessment or data augmentation. |
| Approach: | They propose a training framework that enables bidirectional refinement between LLMs and embedding models by using task-aware prompts to guide the LLM in generating interpretations for the input texts. |
| Outcome: | Experiments on 14 benchmark datasets across 5 tasks demonstrate the effectiveness of the proposed training framework. |
ULLME: A Unified Framework for Large Language Model Embeddings with Generation-Augmented Learning (2024.emnlp-demo)
Copied to clipboard
| Challenge: | Existing frameworks for large language model embeddings have limited support for only a limited range of architectures and fine-tuning strategies. |
| Approach: | They propose a framework that enables bidirectional attention across various LLMs and supports a range of fine-tuning strategies. |
| Outcome: | The proposed framework enables bidirectional attention across various LLMs and supports a range of fine-tuning strategies. |
Large Language Models are Built-in Autoregressive Search Engines (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing dual-encoder dense retrievers obtain representations for questions and documents independently, allowing only shallow interactions between them. |
| Approach: | They propose to use large language models to generate URLs for document retrieval by following human instructions. |
| Outcome: | The proposed method achieves better retrieval performance than existing retrieval approaches on open-domain question answering benchmarks. |
Training LLMs to be Better Text Embedders through Bidirectional Reconstruction (2025.emnlp-main)
Copied to clipboard
Chang Su, Dengliang Shi, Siyuan Huang, Jintao Du, Changhua Meng, Yu Cheng, Weiqiang Wang, Zhouhan Lin
| Challenge: | Existing text embedding approaches often leverage the embeddment of the final token, typically a reserved special token such as ‘[EOS]‘. |
| Approach: | They propose to add a new training stage before contrastive learning to enrich the semantics of the final token embedding. |
| Outcome: | The proposed training stage improves performance on the Massive Text Embedding Benchmark (MTEB), achieving new state-of-the-art results across different LLM base models and scales. |
Causal2Vec: Improving Decoder-only LLMs as Embedding Models through a Contextual Token (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods modify attention mechanism to be bidirectional, undermining LLMs’ ability to extract semantic information acquired during pre-training. |
| Approach: | They propose a general-purpose embedding model that pre-encodes input text into a single Contextual token and then prepends it to the LLM's input sequence. |
| Outcome: | The proposed model improves performance of decoder-only large language models without altering their architectures or introducing significant computational overhead. |
A Text is Worth Several Tokens: Text Embedding from LLMs Secretly Aligns Well with The Key Tokens (2025.acl-long)
Copied to clipboard
| Challenge: | et al., 2023) show that text embeddings from large language models can be aligned with key tokens in input text. |
| Approach: | They propose a sparse retrieval method based on aligned tokens for large language models . they show that this phenomenon is universal and is not affected by model architecture . |
| Outcome: | The proposed method can achieve 80% of the dense retrieval effect of the same model while reducing the computation significantly. |