Papers with decoding step
OPT-Tree: Speculative Decoding with Adaptive Draft Tree Structure (2025.tacl-1)
Copied to clipboard
| Challenge: | Autoregressive language models generate one token in one step, limiting inference efficiency . Existing methods do not adapt to different situations to maximize acceptance length . speculative decoding has shown great potential for lossless acceleration . |
| Approach: | They propose an algorithm to construct adaptive and scalable draft trees for autoregressive language models. |
| Outcome: | Experimental results show that OPT-Tree outperforms existing draft trees and achieves speed-up ratio of up to 3.2 compared with autoregressive decoding. |
DEED: Dynamic Early Exit on Decoder for Accelerating Encoder-Decoder Transformer Models (2024.findings-naacl)
Copied to clipboard
| Challenge: | Encoder-decoder transformer models suffer from high inference latency due to auto-regressive decoding . Typically, the decoder takes up most of the latency because of the auto-decoding - a problem that is not solved by the current model. |
| Approach: | They propose an approach to perform Dynamic Early Exit on Decoder to reduce inference latency by 20%-74% by using a multi-exit encoder-decoder transformer model trained with deep supervision. |
| Outcome: | The proposed model reduces inference latency by 20%-74% with comparable or even higher accuracy compared to baseline models. |
Real-World Compositional Generalization with Disentangled Sequence-to-Sequence Learning (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to compositional generalization have been designed with semantic parsing in mind. |
| Approach: | They propose a disentangled sequence-to-sequence model which encourages more disentanglement and improves its compute and memory efficiency. |
| Outcome: | The proposed model improves generalization performance across existing tasks and datasets and a new machine translation benchmark. |
Attention Optimization for Abstractive Document Summarization (D19-1)
Copied to clipboard
| Challenge: | Abstractive summarization models require attention to reproduce the most salient information. |
| Approach: | They propose to use local and global variances to augment the vanilla attention model to reproduce the most salient information and avoid repetitions. |
| Outcome: | The proposed attention refinement unit can reproduce the most salient information and avoid repetitions on CNN/Daily Mail dataset. |
A Timestep aware Sentence Embedding and Acme Coverage for Brief but Informative Title Generation (2022.findings-naacl)
Copied to clipboard
| Challenge: | Existing methods for title generation are based on timestep aware sentence embeddings, but they are not effective for generating a title with appropriate information in the content. |
| Approach: | They propose a Timestep aware Sentence Embedding mechanism which refreshes the sentences’ embeddings with corresponding key words in different decoding timesteps. |
| Outcome: | The proposed framework outperforms existing methods on various title generation tasks and the evaluation scores are significantly higher than previous approaches. |
Calibrating Structured Output Predictors for Natural Language Processing (2020.acl-main)
Copied to clipboard
| Challenge: | Several modern machine-learning based NLP systems can provide a confidence score with their output predictions. |
| Approach: | They propose a general calibration scheme for output entities of interest in NLP applications that can be used to calibrate confidence scores. |
| Outcome: | The proposed calibration scheme outperforms current calibration techniques for Named Entity Recognition, Part-of-speech tagging and Question Answering systems. |
Iterative GNN-based Decoder for Question Generation (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing models ignore the rich structure information that is hidden in the previously generated text. |
| Approach: | They propose to model the previous generation using a Graph Neural Network at each decoding step. |
| Outcome: | The proposed model outperforms the state-of-the-art models with sentence-level QG tasks on SQUAD and MARCO datasets. |
Double Path Networks for Sequence to Sequence Learning (C18-1)
Copied to clipboard
| Challenge: | Existing approaches for Sequence to Sequence learning have been developed . convolutional neural networks and self-attention networks are the most popular . |
| Approach: | They propose to integrate convolutional and self-attention layers into a double path network for sequence to sequence learning. |
| Outcome: | The proposed method significantly improves performance over state-of-the-art systems. |
Nearest Neighbor Knowledge Distillation for Neural Machine Translation (2022.naacl-main)
Copied to clipboard
| Challenge: | k-nearest-neighbor machine translation (kNN-MT) is a state-of-the-art machine translation technique . however, it requires conducting kNN searches for each decoding step, which increases the cost of decoding . |
| Approach: | They propose to move the time-consuming kNN search forward to the preprocessing phase and introduce k Nearest Neighbor Knowledge Distillation (kNN-KD) that trains the base NMT model to directly learn the knowledge of kN. |
| Outcome: | The proposed method improves over the state-of-the-art model while maintaining the same training and decoding speed as the standard model. |
Unlocking Efficiency in Large Language Model Inference: A Comprehensive Survey of Speculative Decoding (2024.findings-acl)
Copied to clipboard
Heming Xia, Zhe Yang, Qingxiu Dong, Peiyi Wang, Yongqi Li, Tao Ge, Tianyu Liu, Wenjie Li, Zhifang Sui
| Challenge: | Large Language Models (LLMs) have a high inference latency stemming from autoregressive decoding. |
| Approach: | They propose a novel decoding paradigm that drafts multiple tokens and verifies them in parallel . they aim to provide a catalyst for further research on Speculative Decoding . |
| Outcome: | The proposed method drafts multiple tokens and verifies them in parallel . it can be used to accelerate inference in large language models. |
Enhancing Factual Consistency in Text Summarization via Counterfactual Debiasing (2025.coling-main)
Copied to clipboard
| Challenge: | Abstractive text summarization has produced fluent and informative outputs, but factual inconsistency is a challenge. |
| Approach: | They propose a framework that mitigates the causal effects of language bias and irrelevancy bias by counterfactual estimation. |
| Outcome: | The proposed framework outperforms baseline methods on two widely used summarization datasets. |
Towards Opening the Black Box of Neural Machine Translation: Source and Target Interpretations of the Transformer (2022.emnlp-main)
Copied to clipboard
| Challenge: | Neural Machine Translation (NMT) relies on source sentence and target prefix attributions for each input token. |
| Approach: | They propose an interpretability method that tracks input tokens’ attributions for both contexts and extends it to any encoder-decoder Transformer-based model. |
| Outcome: | The proposed method can be extended to any encoder-decoder Transformer-based model and provides insights into their behaviour. |
KCTS: Knowledge-Constrained Tree Search Decoding with Token-Level Hallucination Detection (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies indicate that language models generate non-factual information that is not supported by evidence with a high level of confidence. |
| Approach: | They propose a knowledge-constrained decoding method that guides a frozen LLM to generate text aligned with the reference knowledge at each decoding step. |
| Outcome: | The proposed method reduces the risk of misinformation generated by LLMs by reducing training costs and catastrophic forgetting for multi-tasking models. |
Balancing Diversity and Risk in LLM Sampling: How to Select Your Method and Parameter for Open-Ended Text Generation (2025.acl-long)
Copied to clipboard
| Challenge: | Recent studies suggest that sampling-based decoding strategies can be used to optimize the output of Large Language Models (LLMs) . previous studies have shown that likelihood-maximization produces degenerate text which contains repetitive loops and incoherent context, especially in open-ended tasks. |
| Approach: | They propose to use a prefix tree to estimate the intrinsic capacity of a truncation sampling method by considering the trade-off between diversity and risk at each decoding step. |
| Outcome: | The proposed method is based on a prefix tree which preserves the context of a full sentence. |
Multi-view-guided Passage Reranking with Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing models rely on autoregressive generation and sliding window strategies to rank passages, which incur heavy computational overhead as the number of passages increases. |
| Approach: | They propose a non-generative LLM-based reranking method that encodes query-passage information into diverse view embeddings without being influenced by external biases. |
| Outcome: | The proposed model matches the performance of much larger 7B-scale fine-tuned models while achieving a 100x reduction in inference latency. |