PyTorchKR Papers

최신 논문

총 13,086편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2405.17712

A Context-Aware Approach for Enhancing Data Imputation with Pre-trained Language Models

This paper presents a novel approach named \textbf{C}ontextually \textbf{R}elevant \textbf{I}mputation leveraging pre-trained \textbf{L}anguage \textbf{M}odels (\textbf{CRILM}) for handling missing data in tabular datasets. Instead of relying on traditional numerical estimations, CRILM uses pre-trained language models (LMs) to create contextually relevant descriptors for missing values. This method aligns datasets with LMs' strengths, allowing large LMs to generate these descriptors and small LMs to be fine-tuned on the enriched datasets for enhanced downstream task performance. Our evaluations demonstrate CRILM's superior performance and robustness across MCAR, MAR, and challenging MNAR scenarios, with up to a 10\% improvement over the best-performing baselines. By mitigating biases, particularly in MNAR settings, CRILM improves downstream task performance and offers a cost-effective solution for resource-constrained environments.

arXiv:2405.17627

Salutary Labeling with Zero Human Annotation

Active learning strategically selects informative unlabeled data points and queries their ground truth labels for model training. The prevailing assumption underlying this machine learning paradigm is that acquiring these ground truth labels will optimally enhance model performance. However, this assumption may not always hold true or maximize learning capacity, particularly considering the costly labor annotations required for ground truth labels. In contrast to traditional ground truth labeling, this paper proposes salutary labeling, which automatically assigns the most beneficial labels to the most informative samples without human annotation. Specifically, we utilize the influence function, a tool for estimating sample influence, to select newly added samples and assign their salutary labels by choosing the category that maximizes their positive influence. This process eliminates the need for human annotation. Extensive experiments conducted on nine benchmark datasets demonstrate the superior performance of our salutary labeling approach over traditional active learning strategies. Additionally, we provide several in-depth explorations and practical applications of large language model (LLM) fine-tuning.

arXiv:2405.17534

SMR: State Memory Replay for Long Sequence Modeling

Despite the promising performance of state space models (SSMs) in long sequence modeling, limitations still exist. Advanced SSMs like S5 and S6 (Mamba) in addressing non-uniform sampling, their recursive structures impede efficient SSM computation via convolution. To overcome compatibility limitations in parallel convolutional computation, this paper proposes a novel non-recursive non-uniform sample processing strategy. Theoretical analysis of SSMs through the lens of Event-Triggered Control (ETC) theory reveals the Non-Stable State (NSS) problem, where deviations from sampling point requirements lead to error transmission and accumulation, causing the divergence of the SSM's hidden state. Our analysis further reveals that adjustments of input sequences with early memories can mitigate the NSS problem, achieving Sampling Step Adaptation (SSA). Building on this insight, we introduce a simple yet effective plug-and-play mechanism, State Memory Replay (SMR), which utilizes learnable memories to adjust the current state with multi-step information for generalization at sampling points different from those in the training data. This enables SSMs to stably model varying sampling points. Experiments on long-range modeling tasks in autoregressive language modeling and Long Range Arena demonstrate the general effectiveness of the SMR mechanism for a series of SSM models.

arXiv:2405.18400공식 코드 · Papers with Code

Superposed Decoding: Multiple Generations from a Single Autoregressive Inference Pass

Many applications today provide users with multiple auto-complete drafts as they type, including GitHub's code completion, Gmail's smart compose, and Apple's messaging auto-suggestions. Under the hood, language models support this by running an autoregressive inference pass to provide a draft. Consequently, providing $k$ drafts to the user requires running an expensive language model $k$ times. To alleviate the computation cost of running $k$ inference passes, we propose Superposed Decoding, a new decoding algorithm that generates $k$ drafts at the computation cost of one autoregressive inference pass. We achieve this by feeding a superposition of the most recent token embeddings from the $k$ drafts as input to the next decoding step of the language model. At every inference step we combine the $k$ drafts with the top-$k$ tokens to get $k^2$ new drafts and cache the $k$ most likely options, using an n-gram interpolation with minimal compute overhead to filter out incoherent generations. Our experiments show that $k$ drafts from Superposed Decoding are at least as coherent and factual as Nucleus Sampling and Greedy Decoding respectively, while being at least $2.44\times$ faster for $k\ge3$. In a compute-normalized setting, user evaluations demonstrably favor text generated by Superposed Decoding over Nucleus Sampling. Superposed Decoding can also be combined with other decoding strategies, resulting in universal coverage gains when scaling inference time compute. Code and more examples open-sourced at https://github.com/RAIVNLab/SuperposedDecoding.

arXiv:2405.17815공식 코드 · Papers with Code

Visual Anchors Are Strong Information Aggregators For Multimodal Large Language Model

In the realm of Multimodal Large Language Models (MLLMs), vision-language connector plays a crucial role to link the pre-trained vision encoders with Large Language Models (LLMs). Despite its importance, the vision-language connector has been relatively less explored. In this study, we aim to propose a strong vision-language connector that enables MLLMs to achieve high accuracy while maintain low computation cost. We first reveal the existence of the visual anchors in Vision Transformer and propose a cost-effective search algorithm to extract them. Building on these findings, we introduce the Anchor Former (AcFormer), a novel vision-language connector designed to leverage the rich prior knowledge obtained from these visual anchors during pretraining, guiding the aggregation of information. Through extensive experimentation, we demonstrate that the proposed method significantly reduces computational costs by nearly two-thirds compared with baseline, while simultaneously outperforming baseline methods. This highlights the effectiveness and efficiency of AcFormer. Codes are available at https://github.com/liuhaogeng/Anchor-Former.

arXiv:2405.17728

Facilitating Holistic Evaluations with LLMs: Insights from Scenario-Based Experiments

Workshop courses designed to foster creativity are gaining popularity. However, even experienced faculty teams find it challenging to realize a holistic evaluation that accommodates diverse perspectives. Adequate deliberation is essential to integrate varied assessments, but faculty often lack the time for such exchanges. Deriving an average score without discussion undermines the purpose of a holistic evaluation. Therefore, this paper explores the use of a Large Language Model (LLM) as a facilitator to integrate diverse faculty assessments. Scenario-based experiments were conducted to determine if the LLM could integrate diverse evaluations and explain the underlying pedagogical theories to faculty. The results were noteworthy, showing that the LLM can effectively facilitate faculty discussions. Additionally, the LLM demonstrated the capability to create evaluation criteria by generalizing a single scenario-based experiment, leveraging its already acquired pedagogical domain knowledge.

arXiv:2405.17706공식 코드 · Papers with Code

Video Enriched Retrieval Augmented Generation Using Aligned Video Captions

In this work, we propose the use of "aligned visual captions" as a mechanism for integrating information contained within videos into retrieval augmented generation (RAG) based chat assistant systems. These captions are able to describe the visual and audio content of videos in a large corpus while having the advantage of being in a textual format that is both easy to reason about & incorporate into large language model (LLM) prompts, but also typically require less multimedia content to be inserted into the multimodal LLM context window, where typical configurations can aggressively fill up the context window by sampling video frames from the source video. Furthermore, visual captions can be adapted to specific use cases by prompting the original foundational model / captioner for particular visual details or fine tuning. In hopes of helping advancing progress in this area, we curate a dataset and describe automatic evaluation procedures on common RAG tasks.

arXiv:2405.17637

The Economic Implications of Large Language Model Selection on Earnings and Return on Investment: A Decision Theoretic Model

Selecting language models in business contexts requires a careful analysis of the final financial benefits of the investment. However, the emphasis of academia and industry analysis of LLM is solely on performance. This work introduces a framework to evaluate LLMs, focusing on the earnings and return on investment aspects that should be taken into account in business decision making. We use a decision-theoretic approach to compare the financial impact of different LLMs, considering variables such as the cost per token, the probability of success in the specific task, and the gain and losses associated with LLMs use. The study reveals how the superior accuracy of more expensive models can, under certain conditions, justify a greater investment through more significant earnings but not necessarily a larger RoI. This article provides a framework for companies looking to optimize their technology choices, ensuring that investment in cutting-edge technology aligns with strategic financial objectives. In addition, we discuss how changes in operational variables influence the economics of using LLMs, offering practical insights for enterprise settings, finding that the predicted gain and loss and the different probabilities of success and failure are the variables that most impact the sensitivity of the models.

arXiv:2405.17755

XL3M: A Training-free Framework for LLM Length Extension Based on Segment-wise Inference

Length generalization failure problem, namely the large language model (LLM) fails to generalize to texts longer than its maximum training length, greatly restricts the application of LLM in the scenarios with streaming long inputs. To address this problem, the existing methods either require substantial costs or introduce precision loss. In this paper, we empirically find that the accuracy of the LLM's prediction is highly correlated to its certainty. Based on this, we propose an efficient training free framework, named XL3M (it means extra-long large language model), which enables the LLMs trained on short sequences to reason extremely long sequence without any further training or fine-tuning. Under the XL3M framework, the input context will be firstly decomposed into multiple short sub-contexts, where each sub-context contains an independent segment and a common ``question'' which is a few tokens from the end of the original context. Then XL3M gives a method to measure the relevance between each segment and the ``question'', and constructs a concise key context by splicing all the relevant segments in chronological order. The key context is further used instead of the original context to complete the inference task. Evaluations on comprehensive benchmarks show the superiority of XL3M. Using our framework, a Llama2-7B model is able to reason 20M long sequences on an 8-card Huawei Ascend 910B NPU machine with 64GB memory per card.

arXiv:2405.18414

Don't Forget to Connect! Improving RAG with Graph-based Reranking

Retrieval Augmented Generation (RAG) has greatly improved the performance of Large Language Model (LLM) responses by grounding generation with context from existing documents. These systems work well when documents are clearly relevant to a question context. But what about when a document has partial information, or less obvious connections to the context? And how should we reason about connections between documents? In this work, we seek to answer these two core questions about RAG generation. We introduce G-RAG, a reranker based on graph neural networks (GNNs) between the retriever and reader in RAG. Our method combines both connections between documents and semantic information (via Abstract Meaning Representation graphs) to provide a context-informed ranker for RAG. G-RAG outperforms state-of-the-art approaches while having smaller computational footprint. Additionally, we assess the performance of PaLM 2 as a reranker and find it to significantly underperform G-RAG. This result emphasizes the importance of reranking for RAG even when using Large Language Models.

arXiv:2405.18144공식 코드 · Papers with Code

4-bit Shampoo for Memory-Efficient Network Training

Second-order optimizers, maintaining a matrix termed a preconditioner, are superior to first-order optimizers in both theory and practice. The states forming the preconditioner and its inverse root restrict the maximum size of models trained by second-order optimizers. To address this, compressing 32-bit optimizer states to lower bitwidths has shown promise in reducing memory usage. However, current approaches only pertain to first-order optimizers. In this paper, we propose the first 4-bit second-order optimizers, exemplified by 4-bit Shampoo, maintaining performance similar to that of 32-bit ones. We show that quantizing the eigenvector matrix of the preconditioner in 4-bit Shampoo is remarkably better than quantizing the preconditioner itself both theoretically and experimentally. By rectifying the orthogonality of the quantized eigenvector matrix, we enhance the approximation of the preconditioner's eigenvector matrix, which also benefits the computation of its inverse 4-th root. Besides, we find that linear square quantization slightly outperforms dynamic tree quantization when quantizing second-order optimizer states. Evaluation on various networks for image classification and natural language modeling demonstrates that our 4-bit Shampoo achieves comparable performance to its 32-bit counterpart while being more memory-efficient.

arXiv:2405.18092공식 코드 · Papers with Code

LLM experiments with simulation: Large Language Model Multi-Agent System for Simulation Model Parametrization in Digital Twins

This paper presents a novel design of a multi-agent system framework that applies large language models (LLMs) to automate the parametrization of simulation models in digital twins. This framework features specialized LLM agents tasked with observing, reasoning, decision-making, and summarizing, enabling them to dynamically interact with digital twin simulations to explore parametrization possibilities and determine feasible parameter settings to achieve an objective. The proposed approach enhances the usability of simulation model by infusing it with knowledge heuristics from LLM and enables autonomous search for feasible parametrization to solve a user task. Furthermore, the system has the potential to increase user-friendliness and reduce the cognitive load on human users by assisting in complex decision-making processes. The effectiveness and functionality of the system are demonstrated through a case study, and the visualized demos and codes are available at a GitHub Repository: https://github.com/YuchenXia/LLMDrivenSimulation

arXiv:2405.18064

Automated Real-World Sustainability Data Generation from Images of Buildings

When data on building features is unavailable, the task of determining how to improve that building in terms of carbon emissions becomes infeasible. We show that from only a set of images, a Large Language Model with appropriate prompt engineering and domain knowledge can successfully estimate a range of building features relevant for sustainability calculations. We compare our novel image-to-data method with a ground truth comprising real building data for 47 apartments and achieve accuracy better than a human performing the same task. We also demonstrate that the method can generate tailored recommendations to the owner on how best to improve their properties and discuss methods to scale the approach.

arXiv:2405.17969공식 코드 · Papers with Code

Knowledge Circuits in Pretrained Transformers

The remarkable capabilities of modern large language models are rooted in their vast repositories of knowledge encoded within their parameters, enabling them to perceive the world and engage in reasoning. The inner workings of how these models store knowledge have long been a subject of intense interest and investigation among researchers. To date, most studies have concentrated on isolated components within these models, such as the Multilayer Perceptrons and attention head. In this paper, we delve into the computation graph of the language model to uncover the knowledge circuits that are instrumental in articulating specific knowledge. The experiments, conducted with GPT2 and TinyLLAMA, have allowed us to observe how certain information heads, relation heads, and Multilayer Perceptrons collaboratively encode knowledge within the model. Moreover, we evaluate the impact of current knowledge editing techniques on these knowledge circuits, providing deeper insights into the functioning and constraints of these editing methodologies. Finally, we utilize knowledge circuits to analyze and interpret language model behaviors such as hallucinations and in-context learning. We believe the knowledge circuits hold potential for advancing our understanding of Transformers and guiding the improved design of knowledge editing. Code and data are available in https://github.com/zjunlp/KnowledgeCircuits.

arXiv:2405.17956

Unified Preference Optimization: Language Model Alignment Beyond the Preference Frontier

For aligning large language models (LLMs), prior work has leveraged reinforcement learning via human feedback (RLHF) or variations of direct preference optimization (DPO). While DPO offers a simpler framework based on maximum likelihood estimation, it compromises on the ability to easily tune language models to maximize auxiliary, non-preferential objectives according to the LLM designer's preferences (e.g., tuning lexical style or minimizing specific kinds of harmful content). Critically, these designer objectives may not be amply human-labeled or represented in available data, align with user preferences, or even be able to be captured tractably by binary preference pairs. To leverage the simplicity and performance of DPO with the generality of RL, we propose a unified approach. Based on a simple decomposition of preference and auxiliary objectives, we allow for tuning LLMs to optimize user and designer preferences without any additional specialized or preference data, computational cost, stability ``tweaks'', or training instability. The proposed method, Unified Preference Optimization, shows the ability to effectively generalize to user preferences and auxiliary objectives, while preserving or surpassing alignment performance on challenging benchmarks across a range of model sizes.

arXiv:2405.17093공식 코드 · Papers with Code

DeeperImpact: Optimizing Sparse Learned Index Structures

A lot of recent work has focused on sparse learned indexes that use deep neural architectures to significantly improve retrieval quality while keeping the efficiency benefits of the inverted index. While such sparse learned structures achieve effectiveness far beyond those of traditional inverted index-based rankers, there is still a gap in effectiveness to the best dense retrievers, or even to sparse methods that leverage more expensive optimizations such as query expansion and query term weighting. We focus on narrowing this gap by revisiting and optimizing DeepImpact, a sparse retrieval approach that uses DocT5Query for document expansion followed by a BERT language model to learn impact scores for document terms. We first reinvestigate the expansion process and find that the recently proposed Doc2Query -- query filtration does not enhance retrieval quality when used with DeepImpact. Instead, substituting T5 with a fine-tuned Llama 2 model for query prediction results in a considerable improvement. Subsequently, we study training strategies that have proven effective for other models, in particular the use of hard negatives, distillation, and pre-trained CoCondenser model initialization. Our results substantially narrow the effectiveness gap with the most effective versions of SPLADE.

arXiv:2405.17383

Unlocking the Secrets of Linear Complexity Sequence Model from A Unified Perspective

We present the Linear Complexity Sequence Model (LCSM), a comprehensive solution that unites various sequence modeling techniques with linear complexity, including linear attention, state space model, long convolution, and linear RNN, within a single framework. The goal is to enhance comprehension of these models by analyzing the impact of each component from a cohesive and streamlined viewpoint. Specifically, we segment the modeling processes of these models into three distinct stages: Expand, Oscillation, and Shrink (EOS), with each model having its own specific settings. The Expand stage involves projecting the input signal onto a high-dimensional memory state. This is followed by recursive operations performed on the memory state in the Oscillation stage. Finally, the memory state is projected back to a low-dimensional space in the Shrink stage. We perform comprehensive experiments to analyze the impact of different stage settings on language modeling and retrieval tasks. Our results show that data-driven methods are crucial for the effectiveness of the three stages in language modeling, whereas hand-crafted methods yield better performance in retrieval tasks.

arXiv:2405.17381공식 코드 · Papers with Code

Various Lengths, Constant Speed: Efficient Language Modeling with Lightning Attention

We present Lightning Attention, the first linear attention implementation that maintains a constant training speed for various sequence lengths under fixed memory consumption. Due to the issue with cumulative summation operations (cumsum), previous linear attention implementations cannot achieve their theoretical advantage in a casual setting. However, this issue can be effectively solved by utilizing different attention calculation strategies to compute the different parts of attention. Specifically, we split the attention calculation into intra-blocks and inter-blocks and use conventional attention computation for intra-blocks and linear attention kernel tricks for inter-blocks. This eliminates the need for cumsum in the linear attention calculation. Furthermore, a tiling technique is adopted through both forward and backward procedures to take full advantage of the GPU hardware. To enhance accuracy while preserving efficacy, we introduce TransNormerLLM (TNL), a new architecture that is tailored to our lightning attention. We conduct rigorous testing on standard and self-collected datasets with varying model sizes and sequence lengths. TNL is notably more efficient than other language models. In addition, benchmark results indicate that TNL performs on par with state-of-the-art LLMs utilizing conventional transformer structures. The source code is released at github.com/OpenNLPLab/TransnormerLLM.

arXiv:2405.17052

SelfCP: Compressing Over-Limit Prompt via the Frozen Large Language Model Itself

Long prompt leads to huge hardware costs when using transformer-based Large Language Models (LLMs). Unfortunately, many tasks, such as summarization, inevitably introduce long documents, and the wide application of in-context learning easily makes the prompt length explode. This paper proposes a Self-Compressor (SelfCP), which employs the target LLM itself to compress over-limit prompts into dense vectors while keeping the allowed prompts unmodified. Dense vectors are then projected into dense tokens via a learnable connector to make the same LLM unburden to understand. The connector is supervised-tuned under the language modeling objective of the LLM on relatively long texts selected from publicly accessed datasets, involving an instruction dataset to make SelfCP respond to various prompts, while the target LLM keeps frozen during training. We build the lightweight SelfCP upon 2 different backbones with merely 17M learnable parameters originating from the connector and a learnable embedding. Evaluation on both English and Chinese benchmarks demonstrate that SelfCP effectively substitutes 12$\times$ over-limit prompts with dense tokens to reduce memory costs and booster inference throughputs, yet improving response quality. The outstanding performance brings an efficient solution for LLMs to tackle long prompts without training LLMs from scratch.

arXiv:2405.16533

Chain of Tools: Large Language Model is an Automatic Multi-tool Learner

Augmenting large language models (LLMs) with external tools has emerged as a promising approach to extend their utility, empowering them to solve practical tasks. Existing work typically empowers LLMs as tool users with a manually designed workflow, where the LLM plans a series of tools in a step-by-step manner, and sequentially executes each tool to obtain intermediate results until deriving the final answer. However, they suffer from two challenges in realistic scenarios: (1) The handcrafted control flow is often ad-hoc and constraints the LLM to local planning; (2) The LLM is instructed to use only manually demonstrated tools or well-trained Python functions, which limits its generalization to new tools. In this work, we first propose Automatic Tool Chain (ATC), a framework that enables the LLM to act as a multi-tool user, which directly utilizes a chain of tools through programming. To scale up the scope of the tools, we next propose a black-box probing method. This further empowers the LLM as a tool learner that can actively discover and document tool usages, teaching themselves to properly master new tools. For a comprehensive evaluation, we build a challenging benchmark named ToolFlow, which diverges from previous benchmarks by its long-term planning scenarios and complex toolset. Experiments on both existing datasets and ToolFlow illustrate the superiority of our framework. Analysis on different settings also validates the effectiveness and the utility of our black-box probing algorithm.