Tokenization is a crucial step that bridges human-readable text with model-readable discrete tokens. However, recent studies have revealed that tokenizers can be exploited to elicit unwanted model behaviors. In this work, we investigate incomplete tokens, i.e., undecodable tokens with stray bytes resulting from byte-level byte-pair encoding (BPE) tokenization. We hypothesize that such tokens are heavily reliant on their adjacent tokens and are fragile when paired with unfamiliar tokens. To demonstrate this vulnerability, we introduce improbable bigrams: out-of-distribution combinations of incomplete tokens designed to exploit their dependency. Our experiments show that improbable bigrams are significantly prone to hallucinatory behaviors. Surprisingly, alternative tokenizations of the same phrases result in drastically lower rates of hallucination (93% reduction in Llama3.1). We caution against the potential vulnerabilities introduced by byte-level BPE tokenizers, which may impede the development of trustworthy language models.
相關內容
Answer Set Programming (ASP), a well-known declarative logic programming paradigm, has recently found practical application in Process Mining. In particular, ASP has been used to model tasks involving declarative specifications of business processes. In this area, Declare stands out as the most widely adopted declarative process modeling language, offering a means to model processes through sets of constraints valid traces must satisfy, that can be expressed in Linear Temporal Logic over Finite Traces (LTLf). Existing ASP-based solutions encode Declare constraints by modeling the corresponding LTLf formula or its equivalent automaton which can be obtained using established techniques. In this paper, we introduce a novel encoding for Declare constraints that directly models their semantics as ASP rules, eliminating the need for intermediate representations. We assess the effectiveness of this novel approach on two Process Mining tasks by comparing it with alternative ASP encodings and a Python library for Declare. Under consideration in Theory and Practice of Logic Programming (TPLP).
Knowledge distillation, where a small student model learns from a pre-trained large teacher model, has achieved substantial empirical success since the seminal work of \citep{hinton2015distilling}. Despite prior theoretical studies exploring the benefits of knowledge distillation, an important question remains unanswered: why does soft-label training from the teacher require significantly fewer neurons than directly training a small neural network with hard labels? To address this, we first present motivating experimental results using simple neural network models on a binary classification problem. These results demonstrate that soft-label training consistently outperforms hard-label training in accuracy, with the performance gap becoming more pronounced as the dataset becomes increasingly difficult to classify. We then substantiate these observations with a theoretical contribution based on two-layer neural network models. Specifically, we show that soft-label training using gradient descent requires only $O\left(\frac{1}{\gamma^2 \epsilon}\right)$ neurons to achieve a classification loss averaged over epochs smaller than some $\epsilon > 0$, where $\gamma$ is the separation margin of the limiting kernel. In contrast, hard-label training requires $O\left(\frac{1}{\gamma^4} \cdot \ln\left(\frac{1}{\epsilon}\right)\right)$ neurons, as derived from an adapted version of the gradient descent analysis in \citep{ji2020polylogarithmic}. This implies that when $\gamma \leq \epsilon$, i.e., when the dataset is challenging to classify, the neuron requirement for soft-label training can be significantly lower than that for hard-label training. Finally, we present experimental results on deep neural networks, further validating these theoretical findings.
We introduce the Coarse Payoff-Assessment Learning (CPAL) model, which captures reinforcement learning by boundedly rational decision-makers who focus on the aggregate outcomes of choosing among exogenously defined clusters of alternatives (similarity classes), rather than evaluating each alternative individually. Analyzing a smooth approximation of the model, we show that the learning dynamics exhibit steady-states corresponding to smooth Valuation Equilibria (Jehiel and Samet, 2007). We demonstrate the existence of multiple equilibria in decision trees with generic payoffs and establish the local asymptotic stability of pure equilibria when they occur. Conversely, when trivial choices featuring alternatives within the same similarity class yield sufficiently high payoffs, a unique mixed equilibrium emerges, characterized by indifferences between similarity classes, even under acute sensitivity to payoff differences. Finally, we prove that this unique mixed equilibrium is globally asymptotically stable under the CPAL dynamics.
We introduce a new erasure decoder that applies to arbitrary quantum LDPC codes. Dubbed the cluster decoder, it generalizes the decomposition idea of Vertical-Horizontal (VH) decoding introduced by Connelly et al. in 2022. Like the VH decoder, the idea is to first run the peeling decoder and then post-process the resulting stopping set. The cluster decoder breaks the stopping set into a tree of clusters which can be solved sequentially via Gaussian Elimination (GE). By allowing clusters of unconstrained size, this decoder achieves maximum-likelihood (ML) performance with reduced complexity compared with full GE. When GE is applied only to clusters whose sizes are less than a constant, the performance is degraded but the complexity becomes linear in the block length. Our simulation results show that, for hypergraph product codes, the cluster decoder with constant cluster size achieves near-ML performance similar to VH decoding in the low-erasure-rate regime. For the general quantum LDPC codes we studied, the cluster decoder can be used to estimate the ML performance curve with reduced complexity over a wide range of erasure rates.
In deep learning, the recently introduced state space models utilize HiPPO (High-order Polynomial Projection Operators) memory units to approximate continuous-time trajectories of input functions using ordinary differential equations (ODEs), and these techniques have shown empirical success in capturing long-range dependencies in long input sequences. However, the mathematical foundations of these ODEs, particularly the singular HiPPO-LegS (Legendre Scaled) ODE, and their corresponding numerical discretizations remain unexplored. In this work, we fill this gap by establishing that HiPPO-LegS ODE is well-posed despite its singularity, albeit without the freedom of arbitrary initial conditions, and by establishing convergence of the associated numerical discretization schemes for Riemann-integrable input functions.
Temporal forward-tracking has been the dominant approach for multi-object segmentation and tracking (MOTS). However, a novel time-symmetric tracking methodology has recently been introduced for the detection, segmentation, and tracking of budding yeast cells in pre-recorded samples. Although this architecture has demonstrated a unique perspective on stable and consistent tracking, as well as missed instance re-interpolation, its evaluation has so far been largely confined to settings related to videomicroscopic environments. In this work, we aim to reveal the broader capabilities, advantages, and potential challenges of this architecture across various specifically designed scenarios, including a pedestrian tracking dataset. We also conduct an ablation study comparing the model against its restricted variants and the widely used Kalman filter. Furthermore, we present an attention analysis of the tracking architecture for both pretrained and non-pretrained models
Self-supervised learning, dubbed the dark matter of intelligence, is a promising path to advance machine learning. Yet, much like cooking, training SSL methods is a delicate art with a high barrier to entry. While many components are familiar, successfully training a SSL method involves a dizzying set of choices from the pretext tasks to training hyper-parameters. Our goal is to lower the barrier to entry into SSL research by laying the foundations and latest SSL recipes in the style of a cookbook. We hope to empower the curious researcher to navigate the terrain of methods, understand the role of the various knobs, and gain the know-how required to explore how delicious SSL can be.
Graph neural networks (GNNs) have been demonstrated to be a powerful algorithmic model in broad application fields for their effectiveness in learning over graphs. To scale GNN training up for large-scale and ever-growing graphs, the most promising solution is distributed training which distributes the workload of training across multiple computing nodes. However, the workflows, computational patterns, communication patterns, and optimization techniques of distributed GNN training remain preliminarily understood. In this paper, we provide a comprehensive survey of distributed GNN training by investigating various optimization techniques used in distributed GNN training. First, distributed GNN training is classified into several categories according to their workflows. In addition, their computational patterns and communication patterns, as well as the optimization techniques proposed by recent work are introduced. Second, the software frameworks and hardware platforms of distributed GNN training are also introduced for a deeper understanding. Third, distributed GNN training is compared with distributed training of deep neural networks, emphasizing the uniqueness of distributed GNN training. Finally, interesting issues and opportunities in this field are discussed.
Object detection is a fundamental task in computer vision and image processing. Current deep learning based object detectors have been highly successful with abundant labeled data. But in real life, it is not guaranteed that each object category has enough labeled samples for training. These large object detectors are easy to overfit when the training data is limited. Therefore, it is necessary to introduce few-shot learning and zero-shot learning into object detection, which can be named low-shot object detection together. Low-Shot Object Detection (LSOD) aims to detect objects from a few or even zero labeled data, which can be categorized into few-shot object detection (FSOD) and zero-shot object detection (ZSD), respectively. This paper conducts a comprehensive survey for deep learning based FSOD and ZSD. First, this survey classifies methods for FSOD and ZSD into different categories and discusses the pros and cons of them. Second, this survey reviews dataset settings and evaluation metrics for FSOD and ZSD, then analyzes the performance of different methods on these benchmarks. Finally, this survey discusses future challenges and promising directions for FSOD and ZSD.
Neural machine translation (NMT) is a deep learning based approach for machine translation, which yields the state-of-the-art translation performance in scenarios where large-scale parallel corpora are available. Although the high-quality and domain-specific translation is crucial in the real world, domain-specific corpora are usually scarce or nonexistent, and thus vanilla NMT performs poorly in such scenarios. Domain adaptation that leverages both out-of-domain parallel corpora as well as monolingual corpora for in-domain translation, is very important for domain-specific translation. In this paper, we give a comprehensive survey of the state-of-the-art domain adaptation techniques for NMT.
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