Many self-supervised speech models (S3Ms) have been introduced over the last few years, improving performance and data efficiency on various speech tasks. However, these empirical successes alone do not give a complete picture of what is learned during pre-training. Recent work has begun analyzing how S3Ms encode certain properties, such as phonetic and speaker information, but we still lack a proper understanding of knowledge encoded at the word level and beyond. In this work, we use lightweight analysis methods to study segment-level linguistic properties -- word identity, boundaries, pronunciation, syntactic features, and semantic features -- encoded in S3Ms. We present a comparative study of layer-wise representations from ten S3Ms and find that (i) the frame-level representations within each word segment are not all equally informative, and (ii) the pre-training objective and model size heavily influence the accessibility and distribution of linguistic information across layers. We also find that on several tasks -- word discrimination, word segmentation, and semantic sentence similarity -- S3Ms trained with visual grounding outperform their speech-only counterparts. Finally, our task-based analyses demonstrate improved performance on word segmentation and acoustic word discrimination while using simpler methods than prior work.
相關內容
Instruction-tuned Large Language Models (LLMs) have achieved breakthrough results, opening countless new possibilities for many practical applications. However, LLMs lack elementary safety features that are established norms in other areas of computer science, such as the separation between instructions and data, causing them to malfunction or rendering them vulnerable to manipulation and interference by third parties e.g., via indirect prompt/command injection. Even worse, so far, there is not even an established definition of what precisely such a separation would mean and how its violation could be tested. In this work, we aim to close this gap. We introduce a formal measure to quantify the phenomenon of instruction-data separation as well as an empirical variant of the measure that can be computed from a model`s black-box outputs. We also introduce a new dataset, SEP (Should it be Executed or Processed?), which allows estimating the measure, and we report results on several state-of-the-art open-source and closed LLMs. Finally, we quantitatively demonstrate that all evaluated LLMs fail to achieve a high amount of separation, according to our measure. The source code and SEP dataset are openly accessible at //github.com/egozverev/Shold-It-Be-Executed-Or-Processed.
Large Language Model (LLM) agents have been increasingly adopted as simulation tools to model humans in applications such as social science. However, one fundamental question remains: can LLM agents really simulate human behaviors? In this paper, we focus on one of the most critical behaviors in human interactions, trust, and aim to investigate whether or not LLM agents can simulate human trust behaviors. We first find that LLM agents generally exhibit trust behaviors, referred to as agent trust, under the framework of Trust Games, which are widely recognized in behavioral economics. Then, we discover that LLM agents can have high behavioral alignment with humans regarding trust behaviors, particularly for GPT-4, indicating the feasibility to simulate human trust behaviors with LLM agents. In addition, we probe into the biases in agent trust and the differences in agent trust towards agents and humans. We also explore the intrinsic properties of agent trust under conditions including advanced reasoning strategies and external manipulations. We further offer important implications of our discoveries for various scenarios where trust is paramount. Our study provides new insights into the behaviors of LLM agents and the fundamental analogy between LLMs and humans.
Large language models (LLMs) are increasingly used in robotics, especially for high-level action planning. Meanwhile, many robotics applications involve human supervisors or collaborators. Hence, it is crucial for LLMs to generate socially acceptable actions that align with people's preferences and values. In this work, we test whether LLMs capture people's intuitions about behavior judgments and communication preferences in human-robot interaction (HRI) scenarios. For evaluation, we reproduce three HRI user studies, comparing the output of LLMs with that of real participants. We find that GPT-4 strongly outperforms other models, generating answers that correlate strongly with users' answers in two studies $\unicode{x2014}$ the first study dealing with selecting the most appropriate communicative act for a robot in various situations ($r_s$ = 0.82), and the second with judging the desirability, intentionality, and surprisingness of behavior ($r_s$ = 0.83). However, for the last study, testing whether people judge the behavior of robots and humans differently, no model achieves strong correlations. Moreover, we show that vision models fail to capture the essence of video stimuli and that LLMs tend to rate different communicative acts and behavior desirability higher than people.
As Large Language Models (LLMs) are deployed with increasing real-world responsibilities, it is important to be able to specify and constrain the behavior of these systems in a reliable manner. Model developers may wish to set explicit rules for the model, such as "do not generate abusive content", but these may be circumvented by jailbreaking techniques. Existing evaluations of adversarial attacks and defenses on LLMs generally require either expensive manual review or unreliable heuristic checks. To address this issue, we propose Rule-following Language Evaluation Scenarios (RuLES), a programmatic framework for measuring rule-following ability in LLMs. RuLES consists of 14 simple text scenarios in which the model is instructed to obey various rules while interacting with the user. Each scenario has a programmatic evaluation function to determine whether the model has broken any rules in a conversation. Our evaluations of proprietary and open models show that almost all current models struggle to follow scenario rules, even on straightforward test cases. We also demonstrate that simple optimization attacks suffice to significantly increase failure rates on test cases. We conclude by exploring two potential avenues for improvement: test-time steering and supervised fine-tuning.
Previous face forgery detection methods mainly focus on appearance features, which may be easily attacked by sophisticated manipulation. Considering the majority of current face manipulation methods generate fake faces based on a single frame, which do not take frame consistency and coordination into consideration, artifacts on frame sequences are more effective for face forgery detection. However, current sequence-based face forgery detection methods use general video classification networks directly, which discard the special and discriminative motion information for face manipulation detection. To this end, we propose an effective sequence-based forgery detection framework based on an existing video classification method. To make the motion features more expressive for manipulation detection, we propose an alternative motion consistency block instead of the original motion features module. To make the learned features more generalizable, we propose an auxiliary anomaly detection block. With these two specially designed improvements, we make a general video classification network achieve promising results on three popular face forgery datasets.
Vision-Language Models (VLMs) such as GPT-4V have recently demonstrated incredible strides on diverse vision language tasks. We dig into vision-based deductive reasoning, a more sophisticated but less explored realm, and find previously unexposed blindspots in the current SOTA VLMs. Specifically, we leverage Raven's Progressive Matrices (RPMs), to assess VLMs' abilities to perform multi-hop relational and deductive reasoning relying solely on visual clues. We perform comprehensive evaluations of several popular VLMs employing standard strategies such as in-context learning, self-consistency, and Chain-of-thoughts (CoT) on three diverse datasets, including the Mensa IQ test, IntelligenceTest, and RAVEN. The results reveal that despite the impressive capabilities of LLMs in text-based reasoning, we are still far from achieving comparable proficiency in visual deductive reasoning. We found that certain standard strategies that are effective when applied to LLMs do not seamlessly translate to the challenges presented by visual reasoning tasks. Moreover, a detailed analysis reveals that VLMs struggle to solve these tasks mainly because they are unable to perceive and comprehend multiple, confounding abstract patterns in RPM examples.
We introduce the Approximated Optimal Transport (AOT) technique, a novel training scheme for diffusion-based generative models. Our approach aims to approximate and integrate optimal transport into the training process, significantly enhancing the ability of diffusion models to estimate the denoiser outputs accurately. This improvement leads to ODE trajectories of diffusion models with lower curvature and reduced truncation errors during sampling. We achieve superior image quality and reduced sampling steps by employing AOT in training. Specifically, we achieve FID scores of 1.88 with just 27 NFEs and 1.73 with 29 NFEs in unconditional and conditional generations, respectively. Furthermore, when applying AOT to train the discriminator for guidance, we establish new state-of-the-art FID scores of 1.68 and 1.58 for unconditional and conditional generations, respectively, each with 29 NFEs. This outcome demonstrates the effectiveness of AOT in enhancing the performance of diffusion models.
This study marks a significant advancement by harnessing Large Language Models (LLMs) for multi-intent spoken language understanding (SLU), proposing a unique methodology that capitalizes on the generative power of LLMs within an SLU context. Our innovative technique reconfigures entity slots specifically for LLM application in multi-intent SLU environments and introduces the concept of Sub-Intent Instruction (SII), enhancing the dissection and interpretation of intricate, multi-intent communication within varied domains. The resultant datasets, dubbed LM-MixATIS and LM-MixSNIPS, are crafted from pre-existing benchmarks. Our research illustrates that LLMs can match and potentially excel beyond the capabilities of current state-of-the-art multi-intent SLU models. It further explores LLM efficacy across various intent configurations and dataset proportions. Moreover, we introduce two pioneering metrics, Entity Slot Accuracy (ESA) and Combined Semantic Accuracy (CSA), to provide an in-depth analysis of LLM proficiency in this complex field.
Diffusion models have shown incredible capabilities as generative models; indeed, they power the current state-of-the-art models on text-conditioned image generation such as Imagen and DALL-E 2. In this work we review, demystify, and unify the understanding of diffusion models across both variational and score-based perspectives. We first derive Variational Diffusion Models (VDM) as a special case of a Markovian Hierarchical Variational Autoencoder, where three key assumptions enable tractable computation and scalable optimization of the ELBO. We then prove that optimizing a VDM boils down to learning a neural network to predict one of three potential objectives: the original source input from any arbitrary noisification of it, the original source noise from any arbitrarily noisified input, or the score function of a noisified input at any arbitrary noise level. We then dive deeper into what it means to learn the score function, and connect the variational perspective of a diffusion model explicitly with the Score-based Generative Modeling perspective through Tweedie's Formula. Lastly, we cover how to learn a conditional distribution using diffusion models via guidance.
Embedding models for deterministic Knowledge Graphs (KG) have been extensively studied, with the purpose of capturing latent semantic relations between entities and incorporating the structured knowledge into machine learning. However, there are many KGs that model uncertain knowledge, which typically model the inherent uncertainty of relations facts with a confidence score, and embedding such uncertain knowledge represents an unresolved challenge. The capturing of uncertain knowledge will benefit many knowledge-driven applications such as question answering and semantic search by providing more natural characterization of the knowledge. In this paper, we propose a novel uncertain KG embedding model UKGE, which aims to preserve both structural and uncertainty information of relation facts in the embedding space. Unlike previous models that characterize relation facts with binary classification techniques, UKGE learns embeddings according to the confidence scores of uncertain relation facts. To further enhance the precision of UKGE, we also introduce probabilistic soft logic to infer confidence scores for unseen relation facts during training. We propose and evaluate two variants of UKGE based on different learning objectives. Experiments are conducted on three real-world uncertain KGs via three tasks, i.e. confidence prediction, relation fact ranking, and relation fact classification. UKGE shows effectiveness in capturing uncertain knowledge by achieving promising results on these tasks, and consistently outperforms baselines on these tasks.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191