China

Transformer models face scalability challenges in causal language modeling (CLM) due to inefficient memory allocation for growing key-value (KV) caches, which strains compute and storage resources. Existing methods like Grouped Query Attention (GQA) and token-level KV optimization improve efficiency but rely on rigid resource allocation, often discarding &quot;low-priority&quot; tokens or statically grouping them, failing to address the dynamic spectrum of token importance. We propose mixSGA, a novel mixture-of-expert (MoE) approach that dynamically optimizes token-wise computation and memory allocation. Unlike prior approaches, mixSGA retains all tokens while adaptively routing them to specialized experts with varying KV group sizes, balancing granularity and efficiency. Our key novelties include: (1) a token-wise expert-choice routing mechanism guided by learned importance scores, enabling proportional resource allocation without token discard; (2) weight-sharing across grouped attention projections to minimize parameter overhead; and (3) an auxiliary loss to ensure one-hot routing decisions for training-inference consistency in CLMs. Extensive evaluations across Llama3, TinyLlama, OPT, and Gemma2 model families show mixSGA&#39;s superiority over static baselines. On instruction-following and continued pretraining tasks, mixSGA achieves higher ROUGE-L and lower perplexity under the same KV budgets.

EMNLP 2025

Mixture of Weight-shared Heterogeneous Group Attention Experts for Dynamic Token-wise KV Optimization

kv optimization

heterogeneous moe

efficient lm

It is my great pleasure and honor to welcome you to the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), held in San Diego, California, USA, July 2–7, 2026. ACL 2026 continues our fields tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.

It is exciting to see the many developments in the field reflected at our conference, such as the advancement of large language models (born out of our community) into many application directions. This is not without challenges, partly due to unsolved problems, and partly due to misuse of the technology. All this is reflected well at ACL 2026. The conference also reflects that the field is not only about advancing technology, but also the science of fundamentally understanding language and how it is use by many different speaker communities. 
[Full Message from General Chair Philipp Koehn](https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing) 
<html><button style="display: inline-flex; align-items: center; justify-content: center; white-space: nowrap; border-radius: 9999px; font-weight: bold; background: #7c3aed; color: white; font-family: 'Space Grotesk', sans-serif; height: 40px; font-size: 16px; padding: 0 20px; border: none; cursor: pointer" onclick="window.open('https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing')">Conference Handbook</button></html>

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://2026.aclweb.org/registration/) first.

**Online Registration Form**: https://acl.swoogo.com/acl2026

Registration Required

ACL 2026

The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026) will take place in San Diego, California, from July 2 - 7, 2026.

Transformer models face scalability challenges in causal language modeling (CLM) due to inefficient memory allocation for growing key-value (KV) caches, which strains compute and storage resources. Existing methods like Grouped Query Attention (GQA) and token-level KV optimization improve efficiency but rely on rigid resource allocation, often discarding "low-priority" tokens or statically grouping them, failing to address the dynamic spectrum of token importance. We propose mixSGA, a novel mixture-of-expert (MoE) approach that dynamically optimizes token-wise computation and memory allocation. Unlike prior approaches, mixSGA retains all tokens while adaptively routing them to specialized experts with varying KV group sizes, balancing granularity and efficiency. Our key novelties include: (1) a token-wise expert-choice routing mechanism guided by learned importance scores, enabling proportional resource allocation without token discard; (2) weight-sharing across grouped attention projections to minimize parameter overhead; and (3) an auxiliary loss to ensure one-hot routing decisions for training-inference consistency in CLMs. Extensive evaluations across Llama3, TinyLlama, OPT, and Gemma2 model families show mixSGA's superiority over static baselines. On instruction-following and continued pretraining tasks, mixSGA achieves higher ROUGE-L and lower perplexity under the same KV budgets.

poster

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community." 

*Message from the General Chair, Dirk Hovy*

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

This paper presents R-BPE, a lightweight framework for adapting existing Byte-Pair Encoding (BPE) tokenizers to better support a specified target language. It reuses tokens from user-excluded languages and creates ID-based maps to resolve the new tokens of the chosen language. We evaluate R-BPE on Arabic as a target language. R-BPE reduced subword fertility by an average of 24.4% across the LLaMA 3.1 8B, Command R 35B, and Qwen 3 8B models. Applied to LLaMA 3.1 8B in continued pretraining mode, R-BPE yields a 7.33% reduction in training time. On the ArabicMMLU benchmark, the resulting model improved by 5.09 points on five in-domain topics and matched the original model's overall performance. It also preserved performance on EnglishMMLU. R-BPE effectively leverages existing models' tokenizers, embedding layers, and performance to better support target languages without incurring model size changes. We release an R-BPE implementation that is compatible with HuggingFace interfaces and thereby readily applicable to a wide range of existing models at https://acr.ps/1L9GPmL.

R-BPE: Improving BPE-Tokenizers with Token Reuse

Large Language Models (LLMs) are increasingly costly to fine-tune due to their size, with embedding layers alone accounting for up to 20% of model parameters. While Parameter-Efficient Fine-Tuning (PEFT) methods exist, they largely overlook the embedding layer. In this paper, we introduce TinyTE, a novel PEFT approach that steers model behavior via minimal translational transformations in the embedding space. TinyTE modifies input embeddings without altering hidden layers, achieving achieving competitive performance while requiring approximately 0.0001% of the parameters needed for full fine-tuning. Experiments across architectures provide a new lens for understanding the relationship between input representations and model behavior---revealing them to be more flexible at their foundation than previously thought.

Language Models Can be Efficiently Steered via Minimal Embedding Layer Transformations

Scaling data and model size has driven recent advances in language modeling, but this strategy falters under scenarios with strict data constraints, as in the BabyLM Challenge. However, insights from Chinchilla highlights that smaller models trained on more data outperform larger counterparts trained inadequately, emphasizing the need for compact architectures. Furthermore, while embedding weight tying is a common parameter-saving technique, we find it significantly diminishes linguistic competence in compact models. In response, we explore alternative architectural strategies that preserve the parameter efficiency of tied models without sacrificing the representational benefits of untied embeddings. Consequently, we introduce SLlama a Llama3 architecture variant which incorporates targeted modifications - Repeated Reduced Hidden Size and Projection (RRHP), Permutated Weight Attention (PWA), Shared Projection Multi-Layer Perceptron (SPMLP), and Layer Weight Sharing - to compress transformer components. Without relying on distillation, SLlama achieves a 31.72% improvement in linguistic knowledge acquisition over the BabyLlama baseline, with a comparable GLUE score and significantly lower parameter count. These results demonstrate that well-designed, compact models can rival larger ones under strict data constraints.

SLlama: Parameter-Efficient Language Model Architecture for Enhanced Linguistic Competence Under Strict Data Constraints

Recent advances in large language models (LLMs) have enabled strong reasoning capabilities through Chain-of-Thought (CoT) prompting, which elicits step-by-step problem solving, but often at the cost of excessive verbosity in intermediate outputs, leading to increased computational overhead. We propose Sketch-of-Thought (SoT), a prompting framework that integrates cognitively inspired reasoning paradigms with linguistic constraints to reduce token usage while preserving reasoning accuracy. SoT is designed as a flexible, modular approach and is instantiated with three paradigms—Conceptual Chaining, Chunked Symbolism, and Expert Lexicons—each tailored to distinct reasoning tasks and selected dynamically at test-time by a lightweight routing model. Across 15 reasoning datasets spanning multiple domains, languages, and modalities, SoT achieves token reductions of up to 78% with minimal accuracy loss. In tasks such as mathematical and multi-hop reasoning, it even improves accuracy while shortening outputs.

Sketch-of-Thought: Efficient LLM Reasoning with Adaptive Cognitive-Inspired Sketching

LLM inference for enterprise applications, such as summarization, RAG, and code-generation, typically observe much longer prompt than generations, leading to high prefill cost and response latency. We present SwiftKV, a novel model transformation and distillation procedure targeted at reducing the prefill compute (in FLOPs) of prompt tokens while preserving high generation quality. First, SwiftKV prefills later layers' KV cache using an earlier layer's output, allowing prompt tokens to skip those later layers. Second, SwiftKV employs a lightweight knowledge-preserving distillation procedure that can adapt existing LLMs with minimal accuracy impact. Third, SwiftKV can naturally incorporate KV cache compression to improve inference performance in low-memory scenarios. Our comprehensive experiments show that SwiftKV can effectively reduce prefill computation by 25-50% across several LLM families while incurring minimum quality degradation. In the end-to-end inference serving, SwiftKV realizes up to 2x higher aggregate throughput and 60% lower time per output token. It can achieve a staggering 560 TFlops/GPU of normalized inference throughput, which translates to 16K tokens/s for Llama-3.1-70B. SwiftKV is open-sourced at https://anonymized.link.

SwiftKV: Fast Prefill-Optimized Inference with Knowledge-Preserving Model Transformation

Natural Language Inference (NLI) is a fundamental task in Natural Language Processing (NLP), yet adapting NLI models to new domains remains challenging due to the high cost of collecting domain-specific training data. While prior work proposed 15 sentence transformation rules to automate training data generation, these rules insufficiently capture the diversity of natural language. We propose a novel framework that combines Out-of-Distribution (OOD) detection and BERT-based clustering to identify premise-hypothesis pairs in the SNLI dataset that are not covered by existing rules and to discover four new transformation rules from them. Using these rules with Chain-of-Thought (CoT) prompting and Large Language Models (LLMs), we generate high-quality training data and augment the SNLI dataset. Our method yields consistent performance improvements across dataset sizes, achieving +0.85%p accuracy on 2k and +0.15%p on 550k samples. Furthermore, a distribution-aware augmentation strategy enhances performance across all scales.

Rule Discovery for Natural Language Inference Data Generation Using Out-of-Distribution Detection

Fine-tuning large language models (LLMs) faces significant memory challenges due to the high cost of back-propagation. MeZO addresses this using zeroth-order (ZO) optimization, matching memory usage to inference but suffering from slow convergence due to varying curvatures across model parameters. To overcome this limitation, We propose HELENE, a scalable and memory-efficient optimizer that integrates annealed A-GNB gradients with diagonal Hessian estimation and layer-wise clipping as a second-order pre-conditioner. HELENE provably accelerates and stabilizes convergence by reducing dependence on total parameter space and scaling with the largest layer dimension. Experiments on RoBERTa-large and OPT-1.3B show up to a 20× speedup over MeZO with an average accuracy improvement of 1.5%. HELENE supports full and parameter-efficient fine-tuning, outperforming several state-of-the-art optimizers.

HELENE: Hessian Layer-wise Clipping and Gradient Annealing for Accelerating Fine-tuning LLM with Zeroth-order Optimization

Recent studies have demonstrated that many layers are functionally redundant in large language models (LLMs), enabling model compression by removing these layers to reduce inference cost. While such approaches can improve efficiency, indiscriminate layer pruning often results in significant performance degradation. In this paper, we propose **GRASP** (**G**radient-based **R**etention of **A**daptive **S**ingular **P**arameters), a novel compression framework that mitigates this issue by preserving sensitivity-aware singular values. Unlike direct layer pruning, GRASP leverages gradient-based attribution on a small calibration dataset to adaptively identify and retain critical singular components. By replacing redundant layers with only a minimal set of parameters, GRASP achieves efficient compression while maintaining strong performance with minimal overhead. Experiments across multiple LLMs show that GRASP consistently outperforms existing compression methods, achieving 90% of the original model's performance under a 20% compression ratio.

GRASP: Replace Redundant Layers with Adaptive Singular Parameters for Efficient Model Compression

Vision-Language-Action (VLA) models have made substantial progress by leveraging the robust capabilities of Visual Language Models (VLMs). However, VLMs' significant parameter size and autoregressive (AR) decoding nature impose considerable computational demands on VLA models. While Speculative Decoding (SD) has shown efficacy in accelerating Large Language Models (LLMs) by incorporating efficient drafting and parallel verification, allowing multiple tokens to be generated in one forward pass, its application to VLA models remains unexplored. This work introduces Spec-VLA, an SD framework designed to accelerate VLA models. Due to the difficulty of the action prediction task and the greedy decoding mechanism of the VLA models, the direct application of the advanced SD framework to the VLA prediction task yields a minor speed improvement. To boost the generation speed, we propose an effective mechanism to relax acceptance utilizing the relative distances represented by the action tokens of the VLA model. Empirical results across diverse test scenarios affirm the effectiveness of the Spec-VLA framework, and further analysis substantiates the impact of our proposed strategies, which enhance the acceptance length by 44\%, achieving 1.42times speedup compared with the OpenVLA baseline, without compromising the success rate. The success of the Spec-VLA framework highlights the potential for broader application of speculative execution in VLA prediction scenarios.

Spec-VLA: Speculative Decoding for Vision-Language-Action Models with Relaxed Acceptance

Multilabel text classification is an essential task in NLP applications. Traditional methods require extensive labeled data and are limited to fixed label sets. Extracting labels by LLMs is more effective and universal, but incurs high computational costs. In this work, we introduce a distillation-based T5 generalist model for zero-shot MLTC and few-shot fine-tuning. Our model accommodates variable label sets with general domain-agnostic pertaining, while modeling dependency between labels. Experiments show that our approach outperforms baselines of similar size on three few-shot tasks. Our code is available at https://anonymous.4open.science/r/t5-multilabel-0C32/README.md

Downloads

Next from EMNLP 2025

R-BPE: Improving BPE-Tokenizers with Token Reuse

Similar lecture

Adaptive Layer Selection for Layer-Wise Token Pruning in LLM Inference