Daily Papers

We introduce Guard Vector, a safety task vector computed as the parameter difference between a guardrail model (Guard Model) and a same-architecture pretrained language model. Composing this vector with a target language model yields a Target Guard Model (TGM). We then adapt TGM with a streaming-aware approach that combines prefix-based training and evaluation with a classifier that produces a single-token output. With this composition alone, TGM improves classification quality over established Guard Models across standard safety suites and enables language extensibility to Chinese, Japanese, and Korean, requiring neither additional training nor target language labels. It also demonstrates model portability across two widely used public guardrail backbones, Llama and Gemma. With prefix SFT (supervised fine-tuning), TGM preserves classification quality under streaming by aligning the behavior between prefix inputs and full-text inputs. The single-token output design increases throughput and reduces latency. Together, these components reduce data and compute requirements while promoting streaming-aware evaluation practices, thereby contributing to a more responsible AI ecosystem.

We propose UniTAB that Unifies Text And Box outputs for grounded vision-language (VL) modeling. Grounded VL tasks such as grounded captioning require the model to generate a text description and align predicted words with object regions. To achieve this, models must generate desired text and box outputs together, and meanwhile indicate the alignments between words and boxes. In contrast to existing solutions that use multiple separate modules for different outputs, UniTAB represents both text and box outputs with a shared token sequence, and introduces a special <obj> token to naturally indicate word-box alignments in the sequence. UniTAB thus could provide a more comprehensive and interpretable image description, by freely grounding generated words to object regions. On grounded captioning, UniTAB presents a simpler solution with a single output head, and significantly outperforms state of the art in both grounding and captioning evaluations. On general VL tasks that have different desired output formats (i.e., text, box, or their combination), UniTAB with a single network achieves better or comparable performance than task-specific state of the art. Experiments cover 7 VL benchmarks, including grounded captioning, visual grounding, image captioning, and visual question answering. Furthermore, UniTAB's unified multi-task network and the task-agnostic output sequence design make the model parameter efficient and generalizable to new tasks.

Large Language Models (LLMs) have significantly advanced the field of information retrieval, particularly for reranking. Listwise LLM rerankers have showcased superior performance and generalizability compared to existing supervised approaches. However, conventional listwise LLM reranking methods lack efficiency as they provide ranking output in the form of a generated ordered sequence of candidate passage identifiers. Further, they are trained with the typical language modeling objective, which treats all ranking errors uniformly--potentially at the cost of misranking highly relevant passages. Addressing these limitations, we introduce FIRST, a novel listwise LLM reranking approach leveraging the output logits of the first generated identifier to directly obtain a ranked ordering of the candidates. Further, we incorporate a learning-to-rank loss during training, prioritizing ranking accuracy for the more relevant passages. Empirical results demonstrate that FIRST accelerates inference by 50% while maintaining a robust ranking performance with gains across the BEIR benchmark. Finally, to illustrate the practical effectiveness of listwise LLM rerankers, we investigate their application in providing relevance feedback for retrievers during inference. Our results show that LLM rerankers can provide a stronger distillation signal compared to cross-encoders, yielding substantial improvements in retriever recall after relevance feedback.

We conjecture that hidden state vectors corresponding to individual input tokens encode information sufficient to accurately predict several tokens ahead. More concretely, in this paper we ask: Given a hidden (internal) representation of a single token at position t in an input, can we reliably anticipate the tokens that will appear at positions geq t + 2? To test this, we measure linear approximation and causal intervention methods in GPT-J-6B to evaluate the degree to which individual hidden states in the network contain signal rich enough to predict future hidden states and, ultimately, token outputs. We find that, at some layers, we can approximate a model's output with more than 48% accuracy with respect to its prediction of subsequent tokens through a single hidden state. Finally we present a "Future Lens" visualization that uses these methods to create a new view of transformer states.

The generation quality of large language models (LLMs) is often improved by utilizing inference-time sequence-level scaling methods (e.g., Chain-of-Thought). We introduce hyper-parallel scaling, a complementary framework that improves prediction quality at the token level. Hyper-parallel scaling computes and aggregates multiple output proposals for a single token from the model. We implement this concept in Mixture-of-Experts (MoE) models, which we refer to as Roster of Experts (RoE). RoE is a training-free inference algorithm that turns a single MoE into a dynamic ensemble of MoEs. RoE injects controlled stochasticity into the expert routing mechanism, enabling it to sample multiple diverse experts for each token and aggregate their outputs for a more accurate final prediction.To overcome the computational cost, we introduce an efficient batching strategy and a specialized KV-caching mechanism that minimizes compute and memory overhead. For example, RoE enables a 7B MoE model to match the performance of a 10.5B MoE model while using 30% less compute for inference. These gains are achieved without any fine-tuning of model parameters.

One of the most striking findings in modern research on large language models (LLMs) is that scaling up compute during training leads to better results. However, less attention has been given to the benefits of scaling compute during inference. This survey focuses on these inference-time approaches. We explore three areas under a unified mathematical formalism: token-level generation algorithms, meta-generation algorithms, and efficient generation. Token-level generation algorithms, often called decoding algorithms, operate by sampling a single token at a time or constructing a token-level search space and then selecting an output. These methods typically assume access to a language model's logits, next-token distributions, or probability scores. Meta-generation algorithms work on partial or full sequences, incorporating domain knowledge, enabling backtracking, and integrating external information. Efficient generation methods aim to reduce token costs and improve the speed of generation. Our survey unifies perspectives from three research communities: traditional natural language processing, modern LLMs, and machine learning systems.

Building effective dense retrieval systems remains difficult when relevance supervision is not available. Recent work has looked to overcome this challenge by using a Large Language Model (LLM) to generate hypothetical documents that can be used to find the closest real document. However, this approach relies solely on the LLM to have domain-specific knowledge relevant to the query, which may not be practical. Furthermore, generating hypothetical documents can be inefficient as it requires the LLM to generate a large number of tokens for each query. To address these challenges, we introduce Real Document Embeddings from Relevance Feedback (ReDE-RF). Inspired by relevance feedback, ReDE-RF proposes to re-frame hypothetical document generation as a relevance estimation task, using an LLM to select which documents should be used for nearest neighbor search. Through this re-framing, the LLM no longer needs domain-specific knowledge but only needs to judge what is relevant. Additionally, relevance estimation only requires the LLM to output a single token, thereby improving search latency. Our experiments show that ReDE-RF consistently surpasses state-of-the-art zero-shot dense retrieval methods across a wide range of low-resource retrieval datasets while also making significant improvements in latency per-query.

Since the rise of neural models of code that can generate long expressions and statements rather than a single next-token, one of the major problems has been reliably evaluating their generated output. In this paper, we propose CodeBERTScore: an automatic evaluation metric for code generation, which builds on BERTScore (Zhang et al., 2020). Instead of measuring exact token matching as BLEU, CodeBERTScore computes a soft similarity score between each token in the generated code and in the reference code, using the contextual encodings of large pretrained models. Further, instead of encoding only the generated tokens as in BERTScore, CodeBERTScore also encodes the programmatic context surrounding the generated code. We perform an extensive evaluation of CodeBERTScore across four programming languages. We find that CodeBERTScore achieves a higher correlation with human preference and with functional correctness than all existing metrics. That is, generated code that receives a higher score by CodeBERTScore is more likely to be preferred by humans, as well as to function correctly when executed. Finally, while CodeBERTScore can be used with a multilingual CodeBERT as its base model, we release five language-specific pretrained models to use with our publicly available code at https://github.com/neulab/code-bert-score . Our language-specific models have been downloaded more than 25,000 times from the Huggingface Hub.

In real-world applications, users often require both translations and transcriptions of speech to enhance their comprehension, particularly in streaming scenarios where incremental generation is necessary. This paper introduces a streaming Transformer-Transducer that jointly generates automatic speech recognition (ASR) and speech translation (ST) outputs using a single decoder. To produce ASR and ST content effectively with minimal latency, we propose a joint token-level serialized output training method that interleaves source and target words by leveraging an off-the-shelf textual aligner. Experiments in monolingual (it-en) and multilingual (\{de,es,it\}-en) settings demonstrate that our approach achieves the best quality-latency balance. With an average ASR latency of 1s and ST latency of 1.3s, our model shows no degradation or even improves output quality compared to separate ASR and ST models, yielding an average improvement of 1.1 WER and 0.4 BLEU in the multilingual case.

Unlike language tasks, where the output space is usually limited to a set of tokens, the output space of visual tasks is more complicated, making it difficult to build a unified visual model for various visual tasks. In this paper, we seek to unify the output space of visual tasks, so that we can also build a unified model for visual tasks. To this end, we demonstrate a single unified model that simultaneously handles two typical visual tasks of instance segmentation and depth estimation, which have discrete/fixed-length and continuous/varied-length outputs, respectively. We propose several new techniques that take into account the particularity of visual tasks: 1) Soft token. We employ soft token to represent the task output. Unlike hard tokens in the common VQ-VAE which are assigned one-hot to discrete codebooks/vocabularies, the soft token is assigned softly to the codebook embeddings. Soft token can improve the accuracy of both the next token inference and decoding of the task output; 2) Mask augmentation. Many visual tasks have corruption, undefined or invalid values in label annotations, i.e., occluded area of depth maps. We show that a mask augmentation technique can greatly benefit these tasks. With these new techniques and other designs, we show that the proposed general-purpose task-solver can perform both instance segmentation and depth estimation well. Particularly, we achieve 0.279 RMSE on the specific task of NYUv2 depth estimation, setting a new record on this benchmark. The general-purpose task-solver, dubbed AiT, is available at https://github.com/SwinTransformer/AiT.

In this paper, we propose a unified framework that leverages a single pretrained LLM for Motion-related Multimodal Generation, referred to as MoMug. MoMug integrates diffusion-based continuous motion generation with the model's inherent autoregressive discrete text prediction capabilities by fine-tuning a pretrained LLM. This enables seamless switching between continuous motion output and discrete text token prediction within a single model architecture, effectively combining the strengths of both diffusion- and LLM-based approaches. Experimental results show that, compared to the most recent LLM-based baseline, MoMug improves FID by 38% and mean accuracy across seven metrics by 16.61% on the text-to-motion task. Additionally, it improves mean accuracy across eight metrics by 8.44% on the text-to-motion task. To the best of our knowledge, this is the first approach to integrate diffusion- and LLM-based generation within a single model for motion-related multimodal tasks while maintaining low training costs. This establishes a foundation for future advancements in motion-related generation, paving the way for high-quality yet cost-efficient motion synthesis.

Multimodal large language models (MLLMs) are broadly empowering various fields. Despite their advancements, the explainability of MLLMs remains less explored, hindering deeper understanding, model credibility, and effective visualization. Unlike conventional vision models (e.g., CNNs, ViTs, CLIP) that produce a single output, MLLMs generate sequences of tokens progressively, where each generated token depends on the previous context. Therefore, earlier context tokens can introduce redundant activations that interfere with the explanation of later tokens beyond their original information. Existing studies often overlook this issue, but our observations reveal that these redundant correlations can significantly hurt the reliability of explanations. To address this, we propose an estimated causal inference method to mitigate the interference of context to achieve high-quality MLLM explanation, with a novel rank Gaussian filter to further reduce activation noises. We term this method Token Activation Map (TAM) to highlight the consideration of interactions between tokens. TAM also indicates that it excels at explaining multiple tokens of MLLM, which is different from the Class Activation Map (CAM) for a single prediction. Our TAM method significantly outperforms existing SoTA methods, showcasing high-quality visualization results that can be utilized for various scenarios, such as object localization, failure case analysis, video visualization, MLLMs visual comparison, and model understanding (e.g., color, shape, action, location, visual reasoning, multi-turn conversation, etc). The code is available atgithub.com/xmed-lab/TAM.

Large Multimodal Models (LMMs) such as LLaVA have shown strong performance in visual-linguistic reasoning. These models first embed images into a fixed large number of visual tokens and then feed them into a Large Language Model (LLM). However, this design causes an excessive number of tokens for dense visual scenarios such as high-resolution images and videos, leading to great inefficiency. While token pruning/merging methods do exist, they produce a single length output for each image and do not afford flexibility in trading off information density v.s. efficiency. Inspired by the concept of Matryoshka Dolls, we propose M3: Matryoshka Multimodal Models, which learns to represent visual content as nested sets of visual tokens that capture information across multiple coarse-to-fine granularities. Our approach offers several unique benefits for LMMs: (1) One can explicitly control the visual granularity per test instance during inference, e.g. , adjusting the number of tokens used to represent an image based on the anticipated complexity or simplicity of the content; (2) M3 provides a framework for analyzing the granularity needed for existing datasets, where we find that COCO-style benchmarks only need around ~9 visual tokens to obtain accuracy similar to that of using all 576 tokens; (3) Our approach provides a foundation to explore the best trade-off between performance and visual token length at sample level, where our investigation reveals that a large gap exists between the oracle upper bound and current fixed-scale representations.

Large language models (LLMs) have recently shown strong potential in audio-visual speech recognition (AVSR), but their high computational demands and sensitivity to token granularity limit their practicality in resource-constrained settings. Token compression methods can reduce inference cost, but they require fixing a compression rate in advance and produce a single fixed-length output, offering no flexibility to balance information density and efficiency at inference time. Matryoshka representation learning (MRL) addresses this by enabling a single model to operate across multiple token granularities, allowing compression rates to be adjusted dynamically. However, current MRL-based methods treat each scale independently during training, limiting cross-scale generalization, robustness at high compression, and interpretability. To overcome these limitations, we propose MoME (Mixture of Matryoshka Experts), a novel framework that integrates sparse Mixture-of-Experts (MoE) into MRL-based LLMs for AVSR. MoME augments a frozen LLM with top-k routed and shared experts, allowing dynamic capacity allocation across scales and modalities. A shared router promotes consistent expert activation across granularities, enabling compressed sequences to benefit from representations learned at lower compression. Experiments on LRS2 and LRS3 demonstrate that MoME achieves state-of-the-art performance across AVSR, ASR, and VSR tasks, while requiring significantly fewer parameters and maintaining robustness under noise. MoME unifies the adaptability of MRL with the efficiency of MoE, offering a scalable and interpretable solution for resource-aware speech recognition.

We consider multi-draft speculative sampling, where the proposal sequences are sampled independently from different draft models. At each step, a token-level draft selection scheme takes a list of valid tokens as input and produces an output token whose distribution matches that of the target model. Previous works have demonstrated that the optimal scheme (which maximizes the probability of accepting one of the input tokens) can be cast as a solution to a linear program. In this work we show that the optimal scheme can be decomposed into a two-step solution: in the first step an importance sampling (IS) type scheme is used to select one intermediate token; in the second step (single-draft) speculative sampling is applied to generate the output token. For the case of two identical draft models we further 1) establish a necessary and sufficient condition on the distributions of the target and draft models for the acceptance probability to equal one and 2) provide an explicit expression for the optimal acceptance probability. Our theoretical analysis also motives a new class of token-level selection scheme based on weighted importance sampling. Our experimental results demonstrate consistent improvements in the achievable block efficiency and token rates over baseline schemes in a number of scenarios.

Recent advancements in Optical Character Recognition (OCR) have been driven by transformer-based models. OCR systems are critical in numerous high-stakes domains, yet their vulnerability to adversarial attack remains largely uncharted territory, raising concerns about security and compliance with emerging AI regulations. In this work we present a novel framework to assess the resilience of Transformer-based OCR (TrOCR) models. We develop and assess algorithms for both targeted and untargeted attacks. For the untargeted case, we measure the Character Error Rate (CER), while for the targeted case we use the success ratio. We find that TrOCR is highly vulnerable to untargeted attacks and somewhat less vulnerable to targeted attacks. On a benchmark handwriting data set, untargeted attacks can cause a CER of more than 1 without being noticeable to the eye. With a similar perturbation size, targeted attacks can lead to success rates of around 25% -- here we attacked single tokens, requiring TrOCR to output the tenth most likely token from a large vocabulary.

Small Language Models (SLMs) enable cost-effective, on-device and latency-sensitive AI applications, yet their deployment in Traditional Chinese (TC) remains hindered by token-level instability - models unpredictably emit non-TC characters or code-switch into other languages. We address this practical reliability gap by creating PureTC-1B, a three-stage stabilization pipeline for Llama-3.2-1B-Instruct (an open-weight, instruction-tuned model released by Meta) using parameter-efficient LoRA adapters. Our method combines Continual Pre-Training (CPT) on TC-centric corpora, Supervised Fine-Tuning (SFT) with instruction data, and Direct Preference Optimization (DPO) using TC-adherence preferences to improve monolingual robustness without full-model retraining. On a benchmark designed to simulate real-world usage, PureTC-1B achieves a 51.3% relative reduction (micro-average) in non-TC output tokens versus the base model. On a Named Entity Translation (NET) task, PureTC-1B further reduces incorrect-language tokens by 77.2% relative to Llama-3B and 57.2% relative to Qwen-1.5B, indicating that robust TC adherence is attainable even at the 1B scale. The pipeline is reproducible, adapter-only, and hardware-friendly, offering practitioners a practical recipe to enhance language stability for TC and potentially other non-English languages.

Transformers have recently dominated the ASR field. Although able to yield good performance, they involve an autoregressive (AR) decoder to generate tokens one by one, which is computationally inefficient. To speed up inference, non-autoregressive (NAR) methods, e.g. single-step NAR, were designed, to enable parallel generation. However, due to an independence assumption within the output tokens, performance of single-step NAR is inferior to that of AR models, especially with a large-scale corpus. There are two challenges to improving single-step NAR: Firstly to accurately predict the number of output tokens and extract hidden variables; secondly, to enhance modeling of interdependence between output tokens. To tackle both challenges, we propose a fast and accurate parallel transformer, termed Paraformer. This utilizes a continuous integrate-and-fire based predictor to predict the number of tokens and generate hidden variables. A glancing language model (GLM) sampler then generates semantic embeddings to enhance the NAR decoder's ability to model context interdependence. Finally, we design a strategy to generate negative samples for minimum word error rate training to further improve performance. Experiments using the public AISHELL-1, AISHELL-2 benchmark, and an industrial-level 20,000 hour task demonstrate that the proposed Paraformer can attain comparable performance to the state-of-the-art AR transformer, with more than 10x speedup.

Research has shown the effectiveness of reasoning (e.g., Chain-of-Thought), planning (e.g., SelfAsk), and retrieval augmented generation strategies to improve the performance of Large Language Models (LLMs) on various tasks, such as question answering. However, using a single fixed strategy to answer different kinds of questions is suboptimal in performance and inefficient in terms of generated output tokens and performed retrievals. In our work, we propose a novel technique DyPlan, to induce a dynamic strategy selection process in LLMs, to improve performance and reduce costs in question-answering. DyPlan incorporates an initial decision step to select the most suitable strategy conditioned on the input question and guides the LLM's response generation accordingly. We extend DyPlan to DyPlan-verify, adding an internal verification and correction process to further enrich the generated answer. Experiments on three prominent multi-hop question answering (MHQA) datasets reveal how DyPlan can improve model performance by 7-13% while reducing the cost by 11-32% relative to the best baseline model.

We introduce MAGNeT, a masked generative sequence modeling method that operates directly over several streams of audio tokens. Unlike prior work, MAGNeT is comprised of a single-stage, non-autoregressive transformer. During training, we predict spans of masked tokens obtained from a masking scheduler, while during inference we gradually construct the output sequence using several decoding steps. To further enhance the quality of the generated audio, we introduce a novel rescoring method in which, we leverage an external pre-trained model to rescore and rank predictions from MAGNeT, which will be then used for later decoding steps. Lastly, we explore a hybrid version of MAGNeT, in which we fuse between autoregressive and non-autoregressive models to generate the first few seconds in an autoregressive manner while the rest of the sequence is being decoded in parallel. We demonstrate the efficiency of MAGNeT for the task of text-to-music and text-to-audio generation and conduct an extensive empirical evaluation, considering both objective metrics and human studies. The proposed approach is comparable to the evaluated baselines, while being significantly faster (x7 faster than the autoregressive baseline). Through ablation studies and analysis, we shed light on the importance of each of the components comprising MAGNeT, together with pointing to the trade-offs between autoregressive and non-autoregressive modeling, considering latency, throughput, and generation quality. Samples are available on our demo page https://pages.cs.huji.ac.il/adiyoss-lab/MAGNeT.

Recent advances with large language models (LLM) illustrate their diverse capabilities. We propose a novel algorithm, staged speculative decoding, to accelerate LLM inference in small-batch, on-device scenarios. We address the low arithmetic intensity of small-batch inference by improving upon previous work in speculative decoding. First, we restructure the speculative batch as a tree, which reduces generation costs and increases the expected tokens per batch. Second, we add a second stage of speculative decoding. Taken together, we reduce single-batch decoding latency by 3.16x with a 762M parameter GPT-2-L model while perfectly preserving output quality.

Current state-of-the-art machine translation systems are based on encoder-decoder architectures, that first encode the input sequence, and then generate an output sequence based on the input encoding. Both are interfaced with an attention mechanism that recombines a fixed encoding of the source tokens based on the decoder state. We propose an alternative approach which instead relies on a single 2D convolutional neural network across both sequences. Each layer of our network re-codes source tokens on the basis of the output sequence produced so far. Attention-like properties are therefore pervasive throughout the network. Our model yields excellent results, outperforming state-of-the-art encoder-decoder systems, while being conceptually simpler and having fewer parameters.

Speculative decoding has proven to be an efficient solution to large language model (LLM) inference, where the small drafter predicts future tokens at a low cost, and the target model is leveraged to verify them in parallel. However, most existing works still draft tokens auto-regressively to maintain sequential dependency in language modeling, which we consider a huge computational burden in speculative decoding. We present ParallelSpec, an alternative to auto-regressive drafting strategies in state-of-the-art speculative decoding approaches. In contrast to auto-regressive drafting in the speculative stage, we train a parallel drafter to serve as an efficient speculative model. ParallelSpec learns to efficiently predict multiple future tokens in parallel using a single model, and it can be integrated into any speculative decoding framework that requires aligning the output distributions of the drafter and the target model with minimal training cost. Experimental results show that ParallelSpec accelerates baseline methods in latency up to 62% on text generation benchmarks from different domains, and it achieves 2.84X overall speedup on the Llama-2-13B model using third-party evaluation criteria.

The recent advancements in large language models (LLMs) have been extraordinary, yet the escalating inference costs associated with them present challenges in real-world applications. To address these challenges, we propose a novel approach called Early-exiting Speculative Decoding (EESD) with lossless acceleration. Specifically, EESD utilizes a segment of the LLM to generate draft tokens, incorporating Early-exiting structures after the first N layers. To enhance the quality of draft tokens, a self-distillation method is integrated. This early-exiting design not only reduces deployment and training costs but also significantly accelerates the token generation speed. Moreover, we introduce a novel sampling mechanism that leverages Thompson Sampling to regulate the generation processes, automatically determining the quantity of draft tokens in each round. The original LLM is then employed to validate these draft tokens through a single forward pass, and thus guarantees that the final output text maintains a distribution consistent with vanilla auto-regressive decoding. The experimental results on both 13B and 70B models demonstrate that our approach decodes tokens at a markedly accelerated rate compared to prior methods, showing the effectiveness of our approach.

Long-context language models exhibit impressive performance but remain challenging to deploy due to high GPU memory demands during inference. We propose Memory-efficient Offloaded Mini-sequence Inference (MOM), a method that partitions critical layers into smaller "mini-sequences" and integrates seamlessly with KV cache offloading. Experiments on various Llama, Qwen, and Mistral models demonstrate that MOM reduces peak memory usage by over 50\% on average. On Meta-Llama-3.2-8B, MOM extends the maximum context length from 155k to 455k tokens on a single A100 80GB GPU, while keeping outputs identical and not compromising accuracy. MOM also maintains highly competitive throughput due to minimal computational overhead and efficient last-layer processing. Compared to traditional chunked prefill methods, MOM achieves a 35\% greater context length extension. More importantly, our method drastically reduces prefill memory consumption, eliminating it as the longstanding dominant memory bottleneck during inference. This breakthrough fundamentally changes research priorities, redirecting future efforts from prefill-stage optimizations to improving decode-stage residual KV cache efficiency.

Speculative decoding (SD) accelerates Large Language Model (LLM) generation by using an efficient draft model to propose the next few tokens, which are verified by the LLM in a single forward call, reducing latency while preserving its outputs. We focus on retrieval-based SD where the draft model retrieves the next tokens from a non-parametric datastore. Sparse retrieval (REST), which operates on the surface form of strings, is currently the dominant paradigm due to its simplicity and scalability. However, its effectiveness is limited due to the usage of short contexts and exact string matching. Instead, we introduce Dense Retrieval for Speculative Decoding (DReSD), a novel framework that uses approximate nearest neighbour search with contextualised token embeddings to retrieve the most semantically relevant token sequences for SD. Extensive experiments show that DReSD achieves (on average) 87% higher acceptance rates, 65% longer accepted tokens and 19% faster generation speeds compared to sparse retrieval (REST).