Quantitative Acoustic Microscopy (QAM) is an imaging technology utilising high frequency ultrasound to produce quantitative two-dimensional (2D) maps of acoustical and mechanical properties of biological tissue at microscopy scale. Increased frequency QAM allows for finer resolution at the expense of increased acquisition times and data storage cost. Compressive sampling (CS) methods have been employed to produce QAM images from a reduced sample set, with recent state of the art utilising Approximate Message Passing (AMP) methods. In this paper we investigate the use of AMP-Net, a deep unfolded model for AMP, for the CS reconstruction of QAM parametric maps. Results indicate that AMP-Net can offer superior reconstruction performance even in its stock configuration trained on natural imagery (up to 63% in terms of PSNR), while avoiding the emergence of sampling pattern related artefacts.

This paper introduces *distributed speculative inference (DSI)*, a novel inference algorithm that is provably faster than speculative inference (SI) [leviathan2023, chen2023, miao2024, sun2025, timor2025] and standard autoregressive inference (non-SI). Like other SI algorithms, DSI operates on frozen language models (LMs), requiring no training or architectural modifications, and it preserves the target distribution. Prior studies on SI have demonstrated empirical speedups over non-SI—but rely on sufficiently fast and accurate drafters, which are often unavailable in practice. We identify a gap where SI can be slower than non-SI if drafters are too slow or inaccurate. We close this gap by proving that DSI is faster than both SI and non-SI—given any drafters. DSI is therefore not only faster than SI, but also unlocks the acceleration of LMs for which SI fails. DSI leverages *speculation parallelism (SP)*, a novel type of task parallelism, to orchestrate target and drafter instances that overlap in time, establishing a new foundational tradeoff between computational resources and latency. Our simulations show that DSI is 1.29-1.92x faster than SI in single-node setups for various off-the-shelf LMs and tasks. We open-source all our code.

Accelerating the inference of large language models (LLMs) is a critical challenge in generative AI. Speculative decoding (SD) methods offer substantial efficiency gains by generating multiple tokens using a single target forward pass. However, existing SD approaches require the drafter and target models to share the same vocabulary, thus limiting the pool of possible drafters, often necessitating the training of a drafter from scratch. We present three new SD methods that remove this shared-vocabulary constraint. All three methods preserve the target distribution (i.e., they are lossless) and work with off-the-shelf models without requiring additional training or modifications. Empirically, on summarization, programming, and long-context tasks, our algorithms demonstrate significant speedups of up to 2.8x over standard autoregressive decoding. By enabling any off-the-shelf model to serve as a drafter and requiring no retraining, this work substantially broadens the applicability of the SD framework in practice.

Adaptive inference is a simple method for reducing inference costs.The method works by maintaining multiple classifiers of different capacities, and allocating resources to each test instance according to its difficulty.In this work, we compare the two main approaches for adaptive inference, Early-Exit and Multi-Model, when training data is limited.First, we observe that for models with the same architecture and size, individual Multi-Model classifiers outperform their Early-Exit counterparts by an average of 2.3%.We show that this gap is caused by Early-Exit classifiers sharing model parameters during training, resulting in conflicting gradient updates of model weights.We find that despite this gap, Early-Exit still provides a better speed-accuracy trade-off due to the overhead of the Multi-Model approach.To address these issues, we propose SWEET, 1 an Early-Exit fine-tuning method that assigns each classifier its own set of unique model weights, not updated by other classifiers.We compare SWEET's speed-accuracy curve to standard Early-Exit and Multi-Model baselines and find that it outperforms both methods at fast speeds while maintaining comparable scores to Early-Exit at slow speeds.Moreover, SWEET individual classifiers outperform Early-Exit ones by 1.1% on average.SWEET enjoys the benefits of both methods, paving the way for further reduction of inference costs in NLP.We publicly release our code.2

Paul Roit, Ayal Klein, Daniela Stepanov, Jonathan Mamou, Julian Michael, Gabriel Stanovsky, Luke Zettlemoyer, Ido Dagan. Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics. 2020.

Ayal Klein, Jonathan Mamou, Valentina Pyatkin, Daniela Stepanov, Hangfeng He, Dan Roth, Luke Zettlemoyer, Ido Dagan. Proceedings of the 28th International Conference on Computational Linguistics. 2020.

Deep neural networks (DNNs) have shown much empirical success in solving perceptual tasks across various cognitive modalities. While they are only loosely inspired by the biological brain, recent studies report considerable similarities between representations extracted from task-optimized DNNs and neural populations in the brain. DNNs have subsequently become a popular model class to infer computational principles underlying complex cognitive functions, and in turn, they have also emerged as a natural testbed for applying methods originally developed to probe information in neural populations. In this work, we utilize mean-field theoretic manifold analysis, a recent technique from computational neuroscience that connects geometry of feature representations with linear separability of classes, to analyze language representations from large-scale contextual embedding models. We explore representations from different model families (BERT, RoBERTa, GPT, etc.) and find evidence for emergence of linguistic manifolds across layer depth (e.g., manifolds for part-of-speech tags), especially in ambiguous data (i.e, words with multiple part-of-speech tags, or part-of-speech classes including many words). In addition, we find that the emergence of linear separability in these manifolds is driven by a combined reduction of manifolds’ radius, dimensionality and inter-manifold correlations.

Quantitative acoustic microscopy (QAM) is an imaging modality which uses very-high-frequency ultrasound (i.e., >200 MHz) to form two-dimensional (2D) quantitative images of acoustical and mechanical properties of soft tissues with microscopic resolution (i.e., better than 8 μm). The key component of a QAM system is the ultrasound transducer which must be broadband, have a very small F-number (i.e., <; 1.2), and good sensitivity. In this study, two QAM systems based on a 250-MHz and a 500-MHz transducer are presented, yielding 2D quantitative images at spatial resolution of 7 μm and 4 μm respectively. Thin tissue sections obtained using a microtome or cryotome are raster scanned with precise motors and pulse-echo RF signals are digitized. Inverse models are then used to process each RF signal individually to estimate acoustic impedance, speed of sound, and acoustic attenuation as well as derived parameters such as bulk modulus, mass density, and compressibility. To illustrate the QAM technology and signal processing algorithms, images from cancerous human lymph nodes and ophthalmologic samples are presented and coregistered with histology photomicrographs.

We propose a fully automatic segmentation method called nested graph cut to segment images (2D or 3D) that contain multiple objects with a nested structure. Compared to other graph-cut-based methods developed for multiple regions, our method can work well for nested objects without requiring manual selection of initial seeds, even if different objects have similar intensity distributions and some object boundaries are missing. Promising results were obtained for separating the brain ventricles, the head, and the uterus region in the mouse-embryo head images obtained using high-frequency ultrasound imaging. The proposed method achieved mean Dice similarity coefficients of 0.87 ±0.04 and 0.89 ±0.06 for segmenting BVs and the head, respectively, compared to manual segmentation results by experts on 40 3D images over five gestation stages.

In this paper, we investigate the problem of automatically selecting textual keywords for keyword search development and tuning on audio data for any language. Briefly, the method samples candidate keywords in the training data while trying to match a set of target marginal distributions for keyword features such as keyword frequency in the training or development audio, keyword length, frequency of out-of-vocabulary words, and TF-IDF scores. The method is evaluated on four IARPA Babel program base period languages. We show the use of the automatically selected keywords for the keyword search system development and tuning. We show also that search performance is improved by tuning the decision threshold on the automatically selected keywords.

Keyword search, in the context of low resource languages, has emerged as a key area of research. The dominant approach in keyword search is to use Automatic Speech Recognition (ASR) as a front end to produce a representation of audio that can be indexed. The biggest drawback of this approach lies in its the inability to deal with out-of-vocabulary words and query terms that are not in the ASR system output. In this paper we present an empirical study evaluating various approaches based on using confusion models as query expansion techniques to address this problem. We present results across four languages using a range of confusion models which lead to significant improvements in keyword search performance as measured by the Maximum Term Weighted Value (MTWV) metric.

Automatic speech recognition is a core component of many applications, including keyword search. In this paper we describe experiments on acoustic modeling, language modeling, and decoding for keyword search on a Cantonese conversational telephony corpus collected as part of the IARPA Babel program. We show that acoustic modeling techniques such as the bootstrapped-and-restructured model and deep neural network acoustic model significantly outperform a state-of-the-art baseline GMM/HMM model, in terms of both recognition performance and keyword search performance, with improvements of up to 11% relative character error rate reduction and 31% relative maximum term weighted value improvement. We show that while an interpolated Model M and neural network LM improve recognition performance, they do not improve keyword search results; however, the advanced LM does reduce the size of the keyword search index. Finally, we show that a simple form of automatically adapted keyword search performs 16% better than a preindexed search system, indicating that out-of-vocabulary search is still a challenge.

We present a system for keyword search on Cantonese conversational telephony audio, collected for the IARPA Babel program, that achieves good performance by combining postings lists produced by diverse speech recognition systems from three different research groups. We describe the keyword search task, the data on which the work was done, four different speech recognition systems, and our approach to system combination for keyword search. We show that the combination of four systems outperforms the best single system by 7%, achieving an actual term-weighted value of 0.517.

Spoken content in languages of emerging importance needs to be searchable to provide access to the underlying information. In this paper, we investigate the problem of extending data fusion methodologies from Information Retrieval for Spoken Term Detection on low-resource languages in the framework of the IARPA Babel program. We describe a number of alternative methods improving keyword search performance. We apply these methods to Cantonese, a language that presents some new issues in terms of reduced resources and shorter query lengths. First, we show score normalization methodology that improves in average by 20% keyword search performance. Second, we show that properly combining the outputs of diverse ASR systems performs 14% better than the best normalized ASR system.

The paper presents a new application of automatic speech processing in the Ambient Assisted Living area, developed in the course of a three year research project. Recording and automatic processing of spoken conversations plays a major role in this solution enabling effective search in a personal audio archive and fast browsing of conversations. Processing of elderly conversational speech recorded by a distant PDA microphone poses a great challenge. The speech processing flow includes transcription, speaker tracking and combined indexing and search of spoken terms and participating speakers identity extracted from the audio. We present the entire application and individual speech processing components as well as evaluation results of the individual components and of the end-to-end spoken information retrieval solution.

Spoken queries are a natural medium for searching the Mobile Web. Language modeling for voice search recognition offers different challenges compared to more conventional speech applications. The challenges arise from the fact that spoken queries are usually a set of keywords and do not have a syntactic and grammatical structure. This paper describes a cooccurrence based approach to improve the accuracy of voice queries automatic transcription. With the right choice of scoring function and co-occurrence level, we show that co-occurrence information gives a 2% relative accuracy improvement over a state of the art system.

We present an efficient method for approximate search in a combination of several metric spaces -- which are a generalization of low level image features -- using an inverted index. Our approximation gives very high recall with subsecond response time on a real data set of one million images extracted from Flickr.

We are interested in retrieving information from speech data using phonetic search. We show improvement by expanding the query phonetically using a joint maximum entropy N-gram model. The value of this approach is demonstrated on Broadcast News data from NIST 2006 Spoken Term Detection evaluation.

We are interested in retrieving information from speech data like broadcast news, telephone conversations and roundtable meetings. Today, most systems use large vocabulary continuous speech recognition tools to produce word transcripts; the transcripts are indexed and query terms are retrieved from the index. However, query terms that are not part of the recognizer's vocabulary cannot be retrieved, and the recall of the search is affected. In addition to the output word transcript, advanced systems provide also phonetic transcripts, against which query terms can be matched phonetically. Such phonetic transcripts suffer from lower accuracy and cannot be an alternative to word transcripts.We present a vocabulary independent system that can handle arbitrary queries, exploiting the information provided by having both word transcripts and phonetic transcripts. A speech recognizer generates word confusion networks and phonetic lattices. The transcripts are indexed for query processing and ranking purpose.The value of the proposed method is demonstrated by the relative high performance ofour system, which received the highest overall ranking for US English speech data in the recent NIST Spoken Term Detection evaluation.

We are interested in retrieving information from conversational speech corpora, such as call-center data. This data comprises spontaneous speech conversations with low recording quality, which makes automatic speech recognition (ASR) a highly difficult task. For typical call-center data, even state-of-the-art large vocabulary continuous speech recognition systems produce a transcript with word error rate of 30% or higher. In addition to the output transcript, advanced systems provide word confusion networks (WCNs), a compact representation of word lattices associating each word hypothesis with its posterior probability. Our work exploits the information provided by WCNs in order to improve retrieval performance. In this paper, we show that the mean average precision (MAP) is improved using WCNs compared to the raw word transcripts. Finally, we analyze the effect of increasing ASR word error rate on search effectiveness. We show that MAP is still reasonable even under extremely high error rate.