Bypassing Neural Evaluations for Fast Audio Editing via Adaptive Trajectory Extrapolation

Challenge:	Recent advances in audio diffusion models have significantly improved text-to-audio editing via inversion techniques, but these models typically rely on dense, fixed-step sampling trajectories to maintain structural integrity.
Approach:	They propose a model-agnostic Adaptive Trajectory Extrapolation framework that accelerates inversion-based editing process by dynamically evaluating only the most critical generative phases.
Outcome:	The proposed framework achieves a 3.9 speedup with negligible loss in fidelity.

Similar Papers

Overview

Challenge

Approach

Outcome

From AR to Diffusion: Efficiently Adapting Large Language Models with Strictly Causal and Elastic Horizons (2026.acl-long)

Copied to clipboard

Xiangyu Ma, Teng Xiao, Zuchao Li, Lefei Zhang

Language Models

Challenge:	Autoregressive (AR) models rely on bidirectional attention, creating a structural mismatch with pre-trained Autoregression models.
Approach:	They propose a framework that efficiently adapts autoregressive (AR) models to the diffusion paradigm.
Outcome:	The proposed framework reduces training costs by orders of magnitude while maintaining state-of-the-art performance.

ControlAudio: Tackling Text-Guided, Timing-Indicated and Intelligible Audio Generation via Progressive Diffusion Modeling (2026.acl-long)

Copied to clipboard

Yuxuan Jiang, Zehua Chen, Zeqian Ju, Yusheng Dai, Weibei Dou, Jun Zhu

Speech & Audio in NLP

Challenge:	Recent efforts on text-to-audio generation are exploring fine-grained controllability . however, their performance at scale is limited due to data scarcity .
Approach:	They propose a multi-task learning problem for high-controllability text-to-audio generation . they propose scalable diffusion transformers that augment condition information in sequence .
Outcome:	The proposed method outperforms existing methods on objective and subjective evaluations.

AdaFuse: Adaptive Ensemble Decoding for Large Language Models (2026.acl-long)

Copied to clipboard

Chengming Cui, Tianxin Wei, Ziyi Chen, Ruizhong Qiu, Zhichen Zeng, Zhining Liu, Xuying Ning, Duo Zhou, Jingrui He

Language Models

Challenge:	Existing ensemble approaches to large language models lack flexibility for mid-generation adaptation.
Approach:	They propose an adaptive ensemble decoding framework that dynamically selects semantically appropriate fusion units during generation.
Outcome:	The proposed framework outperforms existing ensemble frameworks on open-domain QA, arithmetic reasoning, and machine translation tasks.

Revisiting Interpolation Augmentation for Speech-to-Text Generation (2024.findings-acl)

Copied to clipboard

Chen Xu, Jie Wang, Xiaoqian Liu, Qian Dong, Chunliang Zhang, Tong Xiao, JingBo Zhu, Dapeng Man, Wu Yang

Text Generation

Challenge:	Existing approaches to speech-to-text generation tasks are limited by the lack of extensive labeled datasets.
Approach:	They propose to use interpolation augmentation to construct virtual training samples by transforming inputs and labels to enhance generalization in other domains.
Outcome:	The proposed approach significantly improves performance across diverse tasks, architectures, and data scales, offering a promising avenue for more robust S2T systems in resource-constrained settings.

Guided by the Plan: Enhancing Faithful Autoregressive Text-to-Audio Generation with Guided Decoding (2026.eacl-long)

Copied to clipboard

Juncheng Wang, Zhe Hu, Chao Xu, Siyue Ren, Yuxiang Feng, Yang Liu, Baigui Sun, Shujun Wang

Multimodality

Challenge:	Autoregressive (AR) models excel at generating temporally coherent audio by producing tokens sequentially, yet they often falter in faithfully following complex textual prompts.
Approach:	They propose a lightweight auxiliary model trained with a GAE-inspired objective to predict final instruction-following quality from partial generations.
Outcome:	The proposed model achieves 10 points improvement in CLAP score over baseline AR models while maintaining computational parity with best-of-N decoding.

FlashAudio: Rectified Flow for Fast and High-Fidelity Text-to-Audio Generation (2025.acl-long)

Copied to clipboard

Huadai Liu, Jialei Wang, Rongjie Huang, Yang Liu, Heng Lu, Zhou Zhao, Wei Xue

Multimodality

Challenge:	Recent advances in latent diffusion models (LDMs) have markedly enhanced text-to-audio generation, yet their iterative sampling processes impose substantial computational demands, limiting practical deployment.
Approach:	They propose to learn straight flow for fast simulation by using flashAudio with rectified flows and immiscible flow to minimize the total distance of data-noise pairs in a batch vias assignment.
Outcome:	The proposed method can learn straight flow for fast simulations and reduce noise distribution.

To Diff or Not to Diff? Structure-Aware and Adaptive Output Formats for Efficient LLM-based Code Editing (2026.findings-acl)

Copied to clipboard

Wei Cheng, Yongchang Cao, Chen Shen, Binhua Li, Jue Chen, Yongbin Li, Wei Hu

Programming Languages in NLP

Challenge:	Large Language Models (LLMs) are increasingly used for code editing, yet the full-code generation paradigm suffers from severe efficiency bottlenecks.
Approach:	They propose to use a structure-aware diff format to train LLMs to choose the most token-efficient format between a given diff format and full code.
Outcome:	The proposed approach matches the most token-efficient format with full-code generation while reducing latency and cost by over 30% on long-code editing tasks.

FiRST: Finetuning Router-Selective Transformers for Input-Adaptive Latency Reduction (2025.findings-emnlp)

Copied to clipboard

Akriti Jain, Saransh Sharma, Koyel Mukherjee, Soumyabrata Pal

Language Models

Challenge:	Existing approaches to improve latency via skipping layers have limitations . fiRST is a model-agnostic framework that reduces inference latency while maintaining quality .
Approach:	They propose a model-agnostic framework that skips transformer layers during decoding . it is fully compatible with KV caching, enabling faster decoding while maintaining quality .
Outcome:	a new framework reduces inference latency by using layer-specific routers to skip transformer layers during decoding.

On the Semantic Latent Space of Diffusion-Based Text-To-Speech Models (2024.acl-short)

Copied to clipboard

Miri Varshavsky-Hassid, Roy Hirsch, Regev Cohen, Tomer Golany, Daniel Freedman, Ehud Rivlin

Multimodality

Challenge:	Denoising Diffusion Models (DDMs) are a powerful generative tool for text-to-speech (TTS) but their semantic capabilities are unknown and control of synthesized speech’s vocal properties remains a challenge.
Approach:	They explore the latent space of frozen TTS models composed of latent bottleneck activations of the DDM’s denoiser and propose methods for finding semantic directions within it.
Outcome:	The proposed methods enable off-the-shelf audio editing without any training, architectural changes or data requirements.

Unlocking Implicit Experience: Synthesizing Tool-Use Trajectories from Text (2026.acl-long)

Copied to clipboard

Zhihao Xu, Rumei Li, Jiahuan Li, Rongxiang Weng, Jingang Wang, Xunliang Cai, Xiting Wang

Information Extraction & Text Mining

Challenge:	Large language models (LLMs) can be used to effectively utilize tools in multi-turn interactions, but acquiring diverse and realistic multi-step tool-use data remains a challenge.
Approach:	They propose a text-based data synthesis pipeline that generates multi-turn tool-use trajectories from text corpora using relevance filtering, workflow tool extraction, trajectory grounding, and complexity refinement.
Outcome:	The proposed model achieves 14.9% improvement on the BFCL V3 Multi-turn benchmark while significantly reducing inference latency and costs.