分享

Plant disease recognition is a critical task that ensures crop health and mitigates the damage caused by diseases. A handy tool that enables farmers to receive a diagnosis based on query pictures or the text description of suspicious plants is in high demand for initiating treatment before potential diseases spread further. In this paper, we develop a multimodal plant disease image retrieval system to support disease search based on either image or text prompts. Specifically, we utilize the largest in-the-wild plant disease dataset PlantWild, which includes over 18,000 images across 89 categories, to provide a comprehensive view of potential diseases relating to the query. Furthermore, cross-modal retrieval is achieved in the developed system, facilitated by a novel CLIP-based vision-language model that encodes both disease descriptions and disease images into the same latent space. Built on top of the retriever, our retrieval system allows users to upload either plant disease images or disease descriptions to retrieve the corresponding images with similar characteristics from the disease dataset to suggest candidate diseases for end users' consideration.

Recent large language model (LLM) defenses have greatly improved models' ability to refuse harmful queries, even when adversarially attacked. However, LLM defenses are primarily evaluated against automated adversarial attacks in a single turn of conversation, an insufficient threat model for real-world malicious use. We demonstrate that multi-turn human jailbreaks uncover significant vulnerabilities, exceeding 70% attack success rate (ASR) on HarmBench against defenses that report single-digit ASRs with automated single-turn attacks. Human jailbreaks also reveal vulnerabilities in machine unlearning defenses, successfully recovering dual-use biosecurity knowledge from unlearned models. We compile these results into Multi-Turn Human Jailbreaks (MHJ), a dataset of 2,912 prompts across 537 multi-turn jailbreaks. We publicly release MHJ alongside a compendium of jailbreak tactics developed across dozens of commercial red teaming engagements, supporting research towards stronger LLM defenses.

We introduce a hierarchical filtering procedure for generative models of sequences on trees, enabling control over the range of positional correlations in the data. Leveraging this controlled setting, we provide evidence that vanilla encoder-only transformer architectures can implement the optimal Belief Propagation algorithm on both root classification and masked language modeling tasks. Correlations at larger distances corresponding to increasing layers of the hierarchy are sequentially included as the network is trained. We analyze how the transformer layers succeed by focusing on attention maps from models trained with varying degrees of filtering. These attention maps show clear evidence for iterative hierarchical reconstruction of correlations, and we can relate these observations to a plausible implementation of the exact inference algorithm for the network sizes considered.

The evolution and diversity of a language is evident from it's various dialects. If the various dialects are not addressed in technological advancements like automatic speech recognition and speech synthesis, there is a chance that these dialects may disappear. Speech technology plays a role in preserving various dialects of a language from going extinct. In order to build a full fledged automatic speech recognition system that addresses various dialects, an Automatic Dialect Identification (ADI) system acting as the front end is required. This is similar to how language identification systems act as front ends to automatic speech recognition systems that handle multiple languages. The current work proposes a way to identify two popular and broadly classified Tamil dialects, namely literary and colloquial Tamil. Acoustical characteristics rather than phonetics and phonotactics are used, alleviating the requirement of language-dependant linguistic tools. Hence one major advantage of the proposed method is that it does not require an annotated corpus, hence it can be easily adapted to other languages. Gaussian Mixture Models (GMM) using Mel Frequency Cepstral Coefficient (MFCC) features are used to perform the classification task. The experiments yielded an error rate of 12%. Vowel nasalization, as being the reason for this good performance, is discussed. The number of mixture models for the GMM is varied and the performance is analysed.

Whisper to normal speech conversion is an active area of research. Various architectures based on generative adversarial networks have been proposed in the recent past. Especially, recent study shows that MaskCycleGAN, which is a mask guided, and cyclic consistency keeping, generative adversarial network, performs really well for voice conversion from spectrogram representations. In the current work we present a MaskCycleGAN approach for the conversion of whispered speech to normal speech. We find that tuning the mask parameters, and pre-processing the signal with a voice activity detector provides superior performance when compared to the existing approach. The wTIMIT dataset is used for evaluation. Objective metrics such as PESQ and G-Loss are used to evaluate the converted speech, along with subjective evaluation using mean opinion score. The results show that the proposed approach offers considerable benefits.

Transformers have evolved with great success in various artificial intelligence tasks. Thanks to our recent prevalence of self-attention mechanisms, which capture long-term dependency, phenomenal outcomes in speech processing and recognition tasks have been produced. The paper presents a comprehensive survey of transformer techniques oriented in speech modality. The main contents of this survey include (1) background of traditional ASR, end-to-end transformer ecosystem, and speech transformers (2) foundational models in a speech via lingualism paradigm, i.e., monolingual, bilingual, multilingual, and cross-lingual (3) dataset and languages, acoustic features, architecture, decoding, and evaluation metric from a specific topological lingualism perspective (4) popular speech transformer toolkit for building end-to-end ASR systems. Finally, highlight the discussion of open challenges and potential research directions for the community to conduct further research in this domain.

Rapid advancements have been made in extending Large Language Models (LLMs) to Large Multi-modal Models (LMMs). However, extending input modality of LLMs to video data remains a challenging endeavor, especially for long videos. Due to insufficient access to large-scale high-quality video data and the excessive compression of visual features, current methods exhibit limitations in effectively processing long videos. In this paper, we introduce Kangaroo, a powerful Video LMM aimed at addressing these challenges. Confronted with issue of inadequate training data, we develop a data curation system to build a large-scale dataset with high-quality annotations for vision-language pre-training and instruction tuning. In addition, we design a curriculum training pipeline with gradually increasing resolution and number of input frames to accommodate long videos. Evaluation results demonstrate that, with 8B parameters, Kangaroo achieves state-of-the-art performance across a variety of video understanding benchmarks while exhibiting competitive results on others. Particularly, on benchmarks specialized for long videos, Kangaroo excels some larger models with over 10B parameters and proprietary models.

Deep speech classification tasks, mainly including keyword spotting and speaker verification, play a crucial role in speech-based human-computer interaction. Recently, the security of these technologies has been demonstrated to be vulnerable to backdoor attacks. Specifically speaking, speech samples are attacked by noisy disruption and component modification in present triggers. We suggest that speech backdoor attacks can strategically focus on emotion, a higher-level subjective perceptual attribute inherent in speech. Furthermore, we proposed that emotional voice conversion technology can serve as the speech backdoor attack trigger, and the method is called EmoAttack. Based on this, we conducted attack experiments on two speech classification tasks, showcasing that EmoAttack method owns impactful trigger effectiveness and its remarkable attack success rate and accuracy variance. Additionally, the ablation experiments found that speech with intensive emotion is more suitable to be targeted for attacks.

Semiconductors, crucial to modern electronics, are generally under-researched in foundational models. It highlights the need for research to enhance the semiconductor device technology portfolio and aid in high-end device fabrication. In this paper, we introduce sLAVA, a small-scale vision-language assistant tailored for semiconductor manufacturing, with a focus on electron microscopy image analysis. It addresses challenges of data scarcity and acquiring high-quality, expert-annotated data. We employ a teacher-student paradigm, using a foundational vision language model like GPT-4 as a teacher to create instruction-following multimodal data for customizing the student model, sLAVA, for electron microscopic image analysis tasks on consumer hardware with limited budgets. Our approach allows enterprises to further fine-tune the proposed framework with their proprietary data securely within their own infrastructure, protecting intellectual property. Rigorous experiments validate that our framework surpasses traditional methods, handles data shifts, and enables high-throughput screening.

This paper focuses on adapting the functionalities of the FastPitch model to the Romanian language; extending the set of speakers from one to eighteen; synthesising speech using an anonymous identity; and replicating the identities of new, unseen speakers. During this work, the effects of various configurations and training strategies were tested and discussed, along with their advantages and weaknesses. Finally, we settled on a new configuration, built on top of the FastPitch architecture, capable of producing natural speech synthesis, for both known (identities from the training dataset) and unknown (identities learnt through short reference samples) speakers. The anonymous speaker can be used for text-to-speech synthesis, if one wants to cancel out the identity information while keeping the semantic content whole and clear. At last, we discussed possible limitations of our work, which will form the basis for future investigations and advancements.

Recently, Text-to-speech (TTS) models based on large language models (LLMs) that translate natural language text into sequences of discrete audio tokens have gained great research attention, with advances in neural audio codec (NAC) models using residual vector quantization (RVQ). However, long-form speech synthesis remains a significant challenge due to the high frame rate, which increases the length of audio tokens and makes it difficult for autoregressive language models to generate audio tokens for even a minute of speech. To address this challenge, this paper introduces two novel post-training approaches: 1) Multi-Resolution Requantization (MReQ) and 2) HALL-E. MReQ is a framework to reduce the frame rate of pre-trained NAC models. Specifically, it incorporates multi-resolution residual vector quantization (MRVQ) module that hierarchically reorganizes discrete audio tokens through teacher-student distillation. HALL-E is an LLM-based TTS model designed to predict hierarchical tokens of MReQ. Specifically, it incorporates the technique of using MRVQ sub-modules and continues training from a pre-trained LLM-based TTS model. Furthermore, to promote TTS research, we create MinutesSpeech, a new benchmark dataset consisting of 40k hours of filtered speech data for training and evaluating speech synthesis ranging from 3s up to 180s. In experiments, we demonstrated the effectiveness of our approaches by applying our post-training framework to VALL-E. We achieved the frame rate down to as low as 8 Hz, enabling the stable minitue-long speech synthesis in a single inference step. Audio samples, dataset, codes and pre-trained models are available at https://yutonishimura-v2.github.io/HALL-E_DEMO/.

The research on audio clue-based target speaker extraction (TSE) has mostly focused on modeling the mixture and reference speech, achieving high performance in English due to the availability of large datasets. However, less attention has been given to the consistent properties of human speech across languages. To bridge this gap, we introduce WHYV (Wanna Hear Your Voice), which addresses the challenge of transferring TSE models from one language to another without fine-tuning. In this work, we proposed a gating mechanism that be able to modify specific frequencies based on the speaker's acoustic features. The model achieves an SI-SDR of 17.3544 on clean English speech and 13.2032 on clean speech mixed with Wham! noise, outperforming all other models in its ability to adapt to different languages.