PyTorchKR Papers

최신 논문

총 13,086편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2406.02120

Diver: Large Language Model Decoding with Span-Level Mutual Information Verification

Large language models (LLMs) have shown impressive capabilities in adapting to various tasks when provided with task-specific instructions. However, LLMs using standard decoding strategies often struggle with deviations from the inputs. Intuitively, compliant LLM outputs should reflect the information present in the input, which can be measured by point-wise mutual information (PMI) scores. Therefore, we propose Diver, a novel approach that enhances LLM Decoding through span-level PMI verification. During inference, Diver first identifies divergence steps that may lead to multiple candidate spans. Subsequently, it calculates the PMI scores by assessing the log-likelihood gains of the input if the candidate spans are generated. Finally, the optimal span is selected based on the PMI re-ranked output distributions. We evaluate our method across various downstream tasks, and empirical results demonstrate that Diver significantly outperforms existing decoding methods in both performance and versatility.

arXiv:2406.02079

Assessing the Performance of Chinese Open Source Large Language Models in Information Extraction Tasks

Information Extraction (IE) plays a crucial role in Natural Language Processing (NLP) by extracting structured information from unstructured text, thereby facilitating seamless integration with various real-world applications that rely on structured data. Despite its significance, recent experiments focusing on English IE tasks have shed light on the challenges faced by Large Language Models (LLMs) in achieving optimal performance, particularly in sub-tasks like Named Entity Recognition (NER). In this paper, we delve into a comprehensive investigation of the performance of mainstream Chinese open-source LLMs in tackling IE tasks, specifically under zero-shot conditions where the models are not fine-tuned for specific tasks. Additionally, we present the outcomes of several few-shot experiments to further gauge the capability of these models. Moreover, our study includes a comparative analysis between these open-source LLMs and ChatGPT, a widely recognized language model, on IE performance. Through meticulous experimentation and analysis, we aim to provide insights into the strengths, limitations, and potential enhancements of existing Chinese open-source LLMs in the domain of Information Extraction within the context of NLP.

arXiv:2406.02009

Phonetic Enhanced Language Modeling for Text-to-Speech Synthesis

Recent language model-based text-to-speech (TTS) frameworks demonstrate scalability and in-context learning capabilities. However, they suffer from robustness issues due to the accumulation of errors in speech unit predictions during autoregressive language modeling. In this paper, we propose a phonetic enhanced language modeling method to improve the performance of TTS models. We leverage self-supervised representations that are phonetically rich as the training target for the autoregressive language model. Subsequently, a non-autoregressive model is employed to predict discrete acoustic codecs that contain fine-grained acoustic details. The TTS model focuses solely on linguistic modeling during autoregressive training, thereby reducing the error propagation that occurs in non-autoregressive training. Both objective and subjective evaluations validate the effectiveness of our proposed method.

arXiv:2406.01983

RKLD: Reverse KL-Divergence-based Knowledge Distillation for Unlearning Personal Information in Large Language Models

With the passage of the Right to Be Forgotten (RTBF) regulations and the scaling up of language model training datasets, research on model unlearning in large language models (LLMs) has become more crucial. Before the era of LLMs, machine unlearning research focused mainly on classification tasks in models with small parameters. In these tasks, the content to be forgotten or retained is clear and straightforward. However, as parameter sizes have grown and tasks have become more complex, balancing forget quality and model utility has become more challenging, especially in scenarios involving personal data instead of classification results. Existing methods based on gradient ascent and its variants often struggle with this balance, leading to unintended information loss or partial forgetting. To address this challenge, we propose RKLD, a novel \textbf{R}everse \textbf{KL}-Divergence-based Knowledge \textbf{D}istillation unlearning algorithm for LLMs targeting the unlearning of personal information. Through RKLD, we achieve significant forget quality and effectively maintain the model utility in our experiments.

arXiv:2406.01976공식 코드 · Papers with Code

Conditional Language Learning with Context

Language models can learn sophisticated language understanding skills from fitting raw text. They also unselectively learn useless corpus statistics and biases, especially during finetuning on domain-specific corpora. In this paper, we propose a simple modification to causal language modeling called conditional finetuning, which performs language modeling conditioned on a context. We show that a context can "explain away" certain corpus statistics and make the model avoid learning them. In this fashion, conditional finetuning achieves selective learning from a corpus, learning knowledge useful for downstream tasks while avoiding learning useless corpus statistics like topic biases. This selective learning effect leads to less forgetting and better stability-plasticity tradeoff in domain finetuning, potentially benefitting lifelong learning with language models.

arXiv:2406.01863

Towards Effective Time-Aware Language Representation: Exploring Enhanced Temporal Understanding in Language Models

In the evolving field of Natural Language Processing, understanding the temporal context of text is increasingly crucial. This study investigates methods to incorporate temporal information during pre-training, aiming to achieve effective time-aware language representation for improved performance on time-related tasks. In contrast to common pre-trained models like BERT, which rely on synchronic document collections such as BookCorpus and Wikipedia, our research introduces BiTimeBERT 2.0, a novel language model pre-trained on a temporal news article collection. BiTimeBERT 2.0 utilizes this temporal news collection, focusing on three innovative pre-training objectives: Time-Aware Masked Language Modeling (TAMLM), Document Dating (DD), and Time-Sensitive Entity Replacement (TSER). Each objective targets a unique aspect of temporal information. TAMLM is designed to enhance the understanding of temporal contexts and relations, DD integrates document timestamps as chronological markers, and TSER focuses on the temporal dynamics of "Person" entities, recognizing their inherent temporal significance. The experimental results consistently demonstrate that BiTimeBERT 2.0 outperforms models like BERT and other existing pre-trained models, achieving substantial gains across a variety of downstream NLP tasks and applications where time plays a pivotal role.

arXiv:2406.01775

OLoRA: Orthonormal Low-Rank Adaptation of Large Language Models

The advent of large language models (LLMs) has revolutionized natural language processing, enabling unprecedented capabilities in understanding and generating human-like text. However, the computational cost and convergence times associated with fine-tuning these models remain significant challenges. Low-Rank Adaptation (LoRA) has emerged as a promising method to mitigate these issues by introducing efficient fine-tuning techniques with a reduced number of trainable parameters. In this paper, we present OLoRA, an enhancement to the LoRA method that leverages orthonormal matrix initialization through QR decomposition. OLoRA significantly accelerates the convergence of LLM training while preserving the efficiency benefits of LoRA, such as the number of trainable parameters and GPU memory footprint. Our empirical evaluations demonstrate that OLoRA not only converges faster but also exhibits improved performance compared to standard LoRA across a variety of language modeling tasks. This advancement opens new avenues for more efficient and accessible fine-tuning of LLMs, potentially enabling broader adoption and innovation in natural language applications.

arXiv:2406.01749공식 코드 · Papers with Code

Towards Harnessing Large Language Models for Comprehension of Conversational Grounding

Conversational grounding is a collaborative mechanism for establishing mutual knowledge among participants engaged in a dialogue. This experimental study analyzes information-seeking conversations to investigate the capabilities of large language models in classifying dialogue turns related to explicit or implicit grounding and predicting grounded knowledge elements. Our experimental results reveal challenges encountered by large language models in the two tasks and discuss ongoing research efforts to enhance large language model-based conversational grounding comprehension through pipeline architectures and knowledge bases. These initiatives aim to develop more effective dialogue systems that are better equipped to handle the intricacies of grounded knowledge in conversations.

arXiv:2406.01633

On Overcoming Miscalibrated Conversational Priors in LLM-based Chatbots

We explore the use of Large Language Model (LLM-based) chatbots to power recommender systems. We observe that the chatbots respond poorly when they encounter under-specified requests (e.g., they make incorrect assumptions, hedge with a long response, or refuse to answer). We conjecture that such miscalibrated response tendencies (i.e., conversational priors) can be attributed to LLM fine-tuning using annotators -- single-turn annotations may not capture multi-turn conversation utility, and the annotators' preferences may not even be representative of users interacting with a recommender system. We first analyze public LLM chat logs to conclude that query under-specification is common. Next, we study synthetic recommendation problems with configurable latent item utilities and frame them as Partially Observed Decision Processes (PODP). We find that pre-trained LLMs can be sub-optimal for PODPs and derive better policies that clarify under-specified queries when appropriate. Then, we re-calibrate LLMs by prompting them with learned control messages to approximate the improved policy. Finally, we show empirically that our lightweight learning approach effectively uses logged conversation data to re-calibrate the response strategies of LLM-based chatbots for recommendation tasks.

arXiv:2406.01306

Unsupervised Distractor Generation via Large Language Model Distilling and Counterfactual Contrastive Decoding

Within the context of reading comprehension, the task of Distractor Generation (DG) aims to generate several incorrect options to confuse readers. Traditional supervised methods for DG rely heavily on expensive human-annotated distractor labels. In this paper, we propose an unsupervised DG framework, leveraging Large Language Models (LLMs) as cost-effective annotators to enhance the DG capability of smaller student models. Specially, to perform knowledge distilling, we propose a dual task training strategy that integrates pseudo distractors from LLMs and the original answer in-formation as the objective targets with a two-stage training process. Moreover, we devise a counterfactual contrastive decoding mechanism for increasing the distracting capability of the DG model. Experiments show that our unsupervised generation method with Bart-base greatly surpasses GPT-3.5-turbo performance with only 200 times fewer model parameters. Our proposed unsupervised DG method offers a cost-effective framework for practical reading comprehension applications, without the need of laborious distractor annotation and costly large-size models

arXiv:2406.01013

Scalable Ensembling For Mitigating Reward Overoptimisation

Reinforcement Learning from Human Feedback (RLHF) has enabled significant advancements within language modeling for powerful, instruction-following models. However, the alignment of these models remains a pressing challenge as the policy tends to overfit the learned ``proxy" reward model past an inflection point of utility as measured by a ``gold" reward model that is more performant -- a phenomenon known as overoptimisation. Prior work has mitigated this issue by computing a pessimistic statistic over an ensemble of reward models, which is common in Offline Reinforcement Learning but incredibly costly for language models with high memory requirements, making such approaches infeasible for sufficiently large models. To this end, we propose using a shared encoder but separate linear heads. We find this leads to similar performance as the full ensemble while allowing tremendous savings in memory and time required for training for models of similar size.

arXiv:2406.00976공식 코드 · Papers with Code

Generative Pre-trained Speech Language Model with Efficient Hierarchical Transformer

While recent advancements in speech language models have achieved significant progress, they face remarkable challenges in modeling the long acoustic sequences of neural audio codecs. In this paper, we introduce \textbf{G}enerative \textbf{P}re-trained \textbf{S}peech \textbf{T}ransformer (GPST), a hierarchical transformer designed for efficient speech language modeling. GPST quantizes audio waveforms into two distinct types of discrete speech representations and integrates them within a hierarchical transformer architecture, allowing for a unified one-stage generation process and enhancing Hi-Res audio generation capabilities. By training on large corpora of speeches in an end-to-end unsupervised manner, GPST can generate syntactically consistent speech with diverse speaker identities. Given a brief 3-second prompt, GPST can produce natural and coherent personalized speech, demonstrating in-context learning abilities. Moreover, our approach can be easily extended to spoken cross-lingual speech generation by incorporating multi-lingual semantic tokens and universal acoustic tokens. Experimental results indicate that GPST significantly outperforms the existing speech language models in terms of word error rate, speech quality, and speaker similarity. The code is available at \url{https://github.com/youngsheen/GPST}.

arXiv:2406.00888공식 코드 · Papers with Code

Aligning Language Models with Demonstrated Feedback

Language models are aligned to emulate the collective voice of many, resulting in outputs that align with no one in particular. Steering LLMs away from generic output is possible through supervised finetuning or RLHF, but requires prohibitively large datasets for new ad-hoc tasks. We argue that it is instead possible to align an LLM to a specific setting by leveraging a very small number (< 10) of demonstrations as feedback. Our method, Demonstration ITerated Task Optimization (DITTO), directly aligns language model outputs to a user's demonstrated behaviors. Derived using ideas from online imitation learning, DITTO cheaply generates online comparison data by treating users' demonstrations as preferred over output from the LLM and its intermediate checkpoints. Concretely, DITTO operates by having an LLM generate examples that are presumed to be inferior to expert demonstrations. The method iteratively constructs pairwise preference relationships between these LLM-generated samples and expert demonstrations, potentially including comparisons between different training checkpoints. These constructed preference pairs are then used to train the model using a preference optimization algorithm (e.g. DPO). We evaluate DITTO's ability to learn fine-grained style and task alignment across domains such as news articles, emails, and blog posts. Additionally, we conduct a user study soliciting a range of demonstrations from participants (N = 16). Across our benchmarks and user study, we find that win-rates for DITTO outperform few-shot prompting, supervised fine-tuning, and other self-play methods by an avg. of 19% points. By using demonstrations as feedback directly, DITTO offers a novel method for effective customization of LLMs.

arXiv:2406.00605

LongSkywork: A Training Recipe for Efficiently Extending Context Length in Large Language Models

We introduce LongSkywork, a long-context Large Language Model (LLM) capable of processing up to 200,000 tokens. We provide a training recipe for efficiently extending context length of LLMs. We identify that the critical element in enhancing long-context processing capability is to incorporate a long-context SFT stage following the standard SFT stage. A mere 200 iterations can convert the standard SFT model into a long-context model. To reduce the effort in collecting and annotating data for long-context language modeling, we develop two novel methods for creating synthetic data. These methods are applied during the continual pretraining phase as well as the Supervised Fine-Tuning (SFT) phase, greatly enhancing the training efficiency of our long-context LLMs. Our findings suggest that synthetic long-context SFT data can surpass the performance of data curated by humans to some extent. LongSkywork achieves outstanding performance on a variety of long-context benchmarks. In the Needle test, a benchmark for long-context information retrieval, our models achieved perfect accuracy across multiple context spans. Moreover, in realistic application scenarios, LongSkywork-13B demonstrates performance on par with Claude2.1, the leading long-context model, underscoring the effectiveness of our proposed methods.

arXiv:2406.00380공식 코드 · Papers with Code

HonestLLM: Toward an Honest and Helpful Large Language Model

Large Language Models (LLMs) have achieved remarkable success across various industries due to their exceptional generative capabilities. However, for safe and effective real-world deployments, ensuring honesty and helpfulness is critical. This paper addresses the question: Can we prioritize the helpfulness of LLMs while preserving their honesty? To begin with, we establish exhaustive principles aimed at guaranteeing the honesty of LLM. Additionally, we introduce a novel dataset, referred to as HoneSet, comprising 930 queries spanning six categories meticulously crafted to assess an LLM's capacity for maintaining honesty. Subsequently, we present two approaches to augmenting honesty and helpfulness in LLMs: a training-free enhancement and a fine-tuning-based improvement. The training-free approach, which is based on curiosity-driven prompting, empowers LLMs to articulate internal confusion and uncertainty regarding queries, thereby optimizing their responses. Conversely, the fine-tuning-based method employs a two-stage process inspired by curriculum learning: initially instructing LLMs to discern between honest and dishonest responses, then refining their training to enhance helpfulness. Experiments conducted on nine prominent LLMs demonstrate a significant improvement in alignment with honesty across all models through the implementation of our proposed enhancements. Particularly noteworthy is the 65.3% enhancement observed in Llama3-8b and the remarkable 124.7% improvement in Mistral-7b, as measured by the H$^{2}$ (honest and helpful) assessment. We believe that our work can pave the way for developing more trustworthy LLMs for real-world applications.

arXiv:2406.00318공식 코드 · Papers with Code

KGLink: A column type annotation method that combines knowledge graph and pre-trained language model

The semantic annotation of tabular data plays a crucial role in various downstream tasks. Previous research has proposed knowledge graph (KG)-based and deep learning-based methods, each with its inherent limitations. KG-based methods encounter difficulties annotating columns when there is no match for column cells in the KG. Moreover, KG-based methods can provide multiple predictions for one column, making it challenging to determine the semantic type with the most suitable granularity for the dataset. This type granularity issue limits their scalability. On the other hand, deep learning-based methods face challenges related to the valuable context missing issue. This occurs when the information within the table is insufficient for determining the correct column type. This paper presents KGLink, a method that combines WikiData KG information with a pre-trained deep learning language model for table column annotation, effectively addressing both type granularity and valuable context missing issues. Through comprehensive experiments on widely used tabular datasets encompassing numeric and string columns with varying type granularity, we showcase the effectiveness and efficiency of KGLink. By leveraging the strengths of KGLink, we successfully surmount challenges related to type granularity and valuable context issues, establishing it as a robust solution for the semantic annotation of tabular data.

arXiv:2406.00244

Controlling Large Language Model Agents with Entropic Activation Steering

The rise of large language models (LLMs) has prompted increasing interest in their use as in-context learning agents. At the core of agentic behavior is the capacity for exploration, or the ability to actively gather information about the environment. But how do LLM agents explore, and how can we control their exploratory behaviors? To answer these questions, we take a representation-level perspective, and introduce Entropic Activation Steering (EAST), an activation steering method for in-context LLM agents. Firstly, we demonstrate that EAST can effectively manipulate an LLM agent's exploration by directly affecting the high-level actions parsed from the outputs of the LLM, in contrast to token-level temperature sampling. Secondly, we reveal how applying this control modulates the uncertainty exhibited in the LLM's thoughts, guiding the agent towards more exploratory actions. Finally, we demonstrate that the steering vectors obtained by EAST generalize across task variants. In total, these results show that LLM agents explicitly encode uncertainty over their actions in their representation space. Our work paves the way for a new understanding of the functioning of LLM agents and to effective control of their decision-making behaviors.

arXiv:2406.02528공식 코드 · Papers with Code

Scalable MatMul-free Language Modeling

Matrix multiplication (MatMul) typically dominates the overall computational cost of large language models (LLMs). This cost only grows as LLMs scale to larger embedding dimensions and context lengths. In this work, we show that MatMul operations can be completely eliminated from LLMs while maintaining strong performance at billion-parameter scales. Our experiments show that our proposed MatMul-free models achieve performance on-par with state-of-the-art Transformers that require far more memory during inference at a scale up to at least 2.7B parameters. We investigate the scaling laws and find that the performance gap between our MatMul-free models and full precision Transformers narrows as the model size increases. We also provide a GPU-efficient implementation of this model which reduces memory usage by up to 61% over an unoptimized baseline during training. By utilizing an optimized kernel during inference, our model's memory consumption can be reduced by more than 10x compared to unoptimized models. To properly quantify the efficiency of our architecture, we build a custom hardware solution on an FPGA which exploits lightweight operations beyond what GPUs are capable of. We processed billion-parameter scale models at 13W beyond human readable throughput, moving LLMs closer to brain-like efficiency. This work not only shows how far LLMs can be stripped back while still performing effectively, but also points at the types of operations future accelerators should be optimized for in processing the next generation of lightweight LLMs. Our code implementation is available at https://github.com/ridgerchu/matmulfreellm.

arXiv:2406.01566공식 코드 · Papers with Code

Helix: Serving Large Language Models over Heterogeneous GPUs and Network via Max-Flow

This paper introduces Helix, a distributed system for high-throughput, low-latency large language model (LLM) serving in heterogeneous GPU clusters. The key idea behind Helix is to formulate inference computation of LLMs over heterogeneous GPUs and network connections as a max-flow problem on directed, weighted graphs, whose nodes represent GPU instances and edges capture both GPU and network heterogeneity through their capacities. Helix then uses a mixed integer linear programming (MILP) algorithm to discover highly optimized strategies to serve LLMs on heterogeneous GPUs. This approach allows Helix to jointly optimize model placement and request scheduling, two highly entangled tasks in heterogeneous LLM serving. Our evaluation on several heterogeneous clusters ranging from 24 to 42 GPU nodes shows that Helix improves serving throughput by up to 3.3x and reduces prompting and decoding latency by up to 66% and 24%, respectively, compared to existing approaches. Helix is available at https://github.com/Thesys-lab/Helix-ASPLOS25.

arXiv:2406.01439

Asynchronous Multi-Server Federated Learning for Geo-Distributed Clients

Federated learning (FL) systems enable multiple clients to train a machine learning model iteratively through synchronously exchanging the intermediate model weights with a single server. The scalability of such FL systems can be limited by two factors: server idle time due to synchronous communication and the risk of a single server becoming the bottleneck. In this paper, we propose a new FL architecture, to our knowledge, the first multi-server FL system that is entirely asynchronous, and therefore addresses these two limitations simultaneously. Our solution keeps both servers and clients continuously active. As in previous multi-server methods, clients interact solely with their nearest server, ensuring efficient update integration into the model. Differently, however, servers also periodically update each other asynchronously, and never postpone interactions with clients. We compare our solution to three representative baselines - FedAvg, FedAsync and HierFAVG - on the MNIST and CIFAR-10 image classification datasets and on the WikiText-2 language modeling dataset. Our solution converges to similar or higher accuracy levels than previous baselines and requires 61% less time to do so in geo-distributed settings.