Large Language Models (LLMs) have demonstrated remarkable capabilities across various fields, from natural language understanding to text generation. Compared to non-generative LLMs like BERT and DeBERTa, generative LLMs like GPT series and Llama series are currently the main focus due to their superior algorithmic performance. The advancements in generative LLMs are closely intertwined with the development of hardware capabilities. Various hardware platforms exhibit distinct hardware characteristics, which can help improve LLM inference performance. Therefore, this paper comprehensively surveys efficient generative LLM inference on different hardware platforms. First, we provide an overview of the algorithm architecture of mainstream generative LLMs and delve into the inference process. Then, we summarize different optimization methods for different platforms such as CPU, GPU, FPGA, ASIC, and PIM/NDP, and provide inference results for generative LLMs. Furthermore, we perform a qualitative and quantitative comparison of inference performance with batch sizes 1 and 8 on different hardware platforms by considering hardware power consumption, absolute inference speed (tokens/s), and energy efficiency (tokens/J). We compare the performance of the same optimization methods across different hardware platforms, the performance across different hardware platforms, and the performance of different methods on the same hardware platform. This provides a systematic and comprehensive summary of existing inference acceleration work by integrating software optimization methods and hardware platforms. We point out that three trends (multimodality, inference-time compute, and higher inference energy efficiency) are promising to redefine the capabilities of edge artificial intelligence systems. Our project is available at https://dai.sjtu.edu.cn/project.html.
Large language models (LLMs), such as GPT series and Llama series have demonstrated strong capabilities in natural language processing, contextual understanding, and text generation. In recent years, researchers are trying to enhance the abilities of LLMs in performing various tasks, and numerous studies have proved that well-designed prompts can significantly improve the performance of LLMs on these tasks. This study compares the language style imitation ability of three different large language models under the guidance of the same zero-shot prompt. It also involves comparing the imitation ability of the same large language model when guided by three different prompts individually. Additionally, by applying a Tree-of-Thoughts (ToT) Prompting method to Llama 3, a conversational AI with the language style of a real person was created. In this study, three evaluation methods were used to evaluate LLMs and prompts. The results show that Llama 3 performs best at imitating language styles, and that the ToT prompting method is the most effective to guide it in imitating language styles. Using a ToT framework, Llama 3 was guided to interact with users in the language style of a specific individual without altering its core parameters, thereby creating a text-based conversational AI that reflects the language style of the individual.
Multimodal Large Language Models (MLLMs) have emerged as a central focus in both industry and academia, but often suffer from biases introduced by visual and language priors, which can lead to multimodal hallucination. These biases arise from the visual encoder and the Large Language Model (LLM) backbone, affecting the attention mechanism responsible for aligning multimodal inputs. Existing decoding-based mitigation methods focus on statistical correlations and overlook the causal relationships between attention mechanisms and model output, limiting their effectiveness in addressing these biases. To tackle this issue, we propose a causal inference framework termed CausalMM that applies structural causal modeling to MLLMs, treating modality priors as a confounder between attention mechanisms and output. Specifically, by employing backdoor adjustment and counterfactual reasoning at both the visual and language attention levels, our method mitigates the negative effects of modality priors and enhances the alignment of MLLM's inputs and outputs, with a maximum score improvement of 65.3% on 6 VLind-Bench indicators and 164 points on MME Benchmark compared to conventional methods. Extensive experiments validate the effectiveness of our approach while being a plug-and-play solution. Our code is available at: https://github.com/The-Martyr/CausalMM
Recently, large language and vision models (LLVMs) have received significant attention and development efforts due to their remarkable generalization performance across a wide range of tasks requiring perception and cognitive abilities. A key factor behind their success is their simple architecture, which consists of a vision encoder, a projector, and a large language model (LLM). Despite their achievements in advanced reasoning tasks, their performance on fundamental perception-related tasks (e.g., MMVP) remains surprisingly low. This discrepancy raises the question of how LLVMs truly perceive images and exploit the advantages of the vision encoder. To address this, we systematically investigate this question regarding several aspects: permutation invariance, robustness, math reasoning, alignment preserving and importance, by evaluating the most common LLVM's families (i.e., LLaVA) across 10 evaluation benchmarks. Our extensive experiments reveal several intriguing properties of current LLVMs: (1) they internally process the image in a global manner, even when the order of visual patch sequences is randomly permuted; (2) they are sometimes able to solve math problems without fully perceiving detailed numerical information; (3) the cross-modal alignment is overfitted to complex reasoning tasks, thereby, causing them to lose some of the original perceptual capabilities of their vision encoder; (4) the representation space in the lower layers (<25%) plays a crucial role in determining performance and enhancing visual understanding. Lastly, based on the above observations, we suggest potential future directions for building better LLVMs and constructing more challenging evaluation benchmarks.
Despite the great success of Large Vision-Language Models (LVLMs), they inevitably suffer from hallucination. As we know, both the visual encoder and the Large Language Model (LLM) decoder in LVLMs are Transformer-based, allowing the model to extract visual information and generate text outputs via attention mechanisms. We find that the attention distribution of LLM decoder on image tokens is highly consistent with the visual encoder and both distributions tend to focus on particular background tokens rather than the referred objects in the image. We attribute to the unexpected attention distribution to an inherent flaw in the visual encoder itself, which misguides LLMs to over emphasize the redundant information and generate object hallucination. To address the issue, we propose DAMRO, a novel training-free strategy that $D$ive into $A$ttention $M$echanism of LVLM to $R$educe $O$bject Hallucination. Specifically, our approach employs classification token (CLS) of ViT to filter out high-attention outlier tokens scattered in the background and then eliminate their influence during decoding stage. We evaluate our method on LVLMs including LLaVA-1.5, LLaVA-NeXT and InstructBLIP, using various benchmarks such as POPE, CHAIR, MME and GPT-4V Aided Evaluation. The results demonstrate that our approach significantly reduces the impact of these outlier tokens, thus effectively alleviating the hallucination of LVLMs. The code of our method will be released soon.
In recent decades, there has been substantial advances in time series models and benchmarks across various individual tasks, such as time series forecasting, classification, and anomaly detection. Meanwhile, compositional reasoning in time series is prevalent in real-world applications (e.g., decision-making and compositional question answering) and is in great demand. Unlike simple tasks that primarily focus on predictive accuracy, compositional reasoning emphasizes the synthesis of diverse information from both time series data and various domain knowledge, making it distinct and extremely more challenging. In this paper, we introduce Compositional Time Series Reasoning, a new task of handling intricate multistep reasoning tasks from time series data. Specifically, this new task focuses on various question instances requiring structural and compositional reasoning abilities on time series data, such as decision-making and compositional question answering. As an initial attempt to tackle this novel task, we developed TS-Reasoner, a program-aided approach that utilizes large language model (LLM) to decompose a complex task into steps of programs that leverage existing time series models and numerical subroutines. Unlike existing reasoning work which only calls off-the-shelf modules, TS-Reasoner allows for the creation of custom modules and provides greater flexibility to incorporate domain knowledge as well as user-specified constraints. We demonstrate the effectiveness of our method through a comprehensive set of experiments. These promising results indicate potential opportunities in the new task of time series reasoning and highlight the need for further research.
Court efficiency is vital for social stability. However, in most countries around the world, the grassroots courts face case backlogs, with decisions relying heavily on judicial personnel's cognitive labor, lacking intelligent tools to improve efficiency. To address this issue, we propose an efficient law article recommendation approach utilizing a Knowledge Graph (KG) and a Large Language Model (LLM). Firstly, we propose a Case-Enhanced Law Article Knowledge Graph (CLAKG) as a database to store current law statutes, historical case information, and correspondence between law articles and historical cases. Additionally, we introduce an automated CLAKG construction method based on LLM. On this basis, we propose a closed-loop law article recommendation method. Finally, through a series of experiments using judgment documents from the website "China Judgements Online", we have improved the accuracy of law article recommendation in cases from 0.549 to 0.694, demonstrating that our proposed method significantly outperforms baseline approaches.
This work presents an interpretable decision-making framework for autonomous vehicles that integrates traffic regulations, norms, and safety guidelines comprehensively and enables seamless adaptation to different regions. While traditional rule-based methods struggle to incorporate the full scope of traffic rules, we develop a Traffic Regulation Retrieval (TRR) Agent based on Retrieval-Augmented Generation (RAG) to automatically retrieve relevant traffic rules and guidelines from extensive regulation documents and relevant records based on the ego vehicle's situation. Given the semantic complexity of the retrieved rules, we also design a reasoning module powered by a Large Language Model (LLM) to interpret these rules, differentiate between mandatory rules and safety guidelines, and assess actions on legal compliance and safety. Additionally, the reasoning is designed to be interpretable, enhancing both transparency and reliability. The framework demonstrates robust performance on both hypothesized and real-world cases across diverse scenarios, along with the ability to adapt to different regions with ease.
Large language models (LLMs) have been used to generate formal proofs of mathematical theorems in proofs assistants such as Lean. However, we often want to optimize a formal proof with respect to various criteria, depending on its downstream use. For example, we may want a proof to adhere to a certain style, or to be readable, concise, or modularly structured. Having suitably optimized proofs is also important for learning tasks, especially since human-written proofs may not optimal for that purpose. To this end, we study a new problem of automated proof optimization: rewriting a proof so that it is correct and optimizes for an arbitrary criterion, such as length or readability. As a first method for automated proof optimization, we present ImProver, a large-language-model agent that rewrites proofs to optimize arbitrary user-defined metrics in Lean. We find that naively applying LLMs to proof optimization falls short, and we incorporate various improvements into ImProver, such as the use of symbolic Lean context in a novel Chain-of-States technique, as well as error-correction and retrieval. We test ImProver on rewriting real-world undergraduate, competition, and research-level mathematics theorems, finding that ImProver is capable of rewriting proofs so that they are substantially shorter, more modular, and more readable.
With the rapid advancement of large language models (LLMs), recent years have witnessed many promising studies on leveraging LLM-based agents to simulate human social behavior. While prior work has demonstrated significant potential across various domains, much of it has focused on specific scenarios involving a limited number of agents and has lacked the ability to adapt when errors occur during simulation. To overcome these limitations, we propose a novel LLM-agent-based simulation platform called \textit{GenSim}, which: (1) \textbf{Abstracts a set of general functions} to simplify the simulation of customized social scenarios; (2) \textbf{Supports one hundred thousand agents} to better simulate large-scale populations in real-world contexts; (3) \textbf{Incorporates error-correction mechanisms} to ensure more reliable and long-term simulations. To evaluate our platform, we assess both the efficiency of large-scale agent simulations and the effectiveness of the error-correction mechanisms. To our knowledge, GenSim represents an initial step toward a general, large-scale, and correctable social simulation platform based on LLM agents, promising to further advance the field of social science.
Data is a crucial element in large language model (LLM) alignment. Recent studies have explored using LLMs for efficient data collection. However, LLM-generated data often suffers from quality issues, with underrepresented or absent aspects and low-quality datapoints. To address these problems, we propose Data Advisor, an enhanced LLM-based method for generating data that takes into account the characteristics of the desired dataset. Starting from a set of pre-defined principles in hand, Data Advisor monitors the status of the generated data, identifies weaknesses in the current dataset, and advises the next iteration of data generation accordingly. Data Advisor can be easily integrated into existing data generation methods to enhance data quality and coverage. Experiments on safety alignment of three representative LLMs (i.e., Mistral, Llama2, and Falcon) demonstrate the effectiveness of Data Advisor in enhancing model safety against various fine-grained safety issues without sacrificing model utility.
Text-to-Music Retrieval, finding music based on a given natural language query, plays a pivotal role in content discovery within extensive music databases. To address this challenge, prior research has predominantly focused on a joint embedding of music audio and text, utilizing it to retrieve music tracks that exactly match descriptive queries related to musical attributes (i.e. genre, instrument) and contextual elements (i.e. mood, theme). However, users also articulate a need to explore music that shares similarities with their favorite tracks or artists, such as \textit{I need a similar track to Superstition by Stevie Wonder}. To address these concerns, this paper proposes an improved Text-to-Music Retrieval model, denoted as TTMR++, which utilizes rich text descriptions generated with a finetuned large language model and metadata. To accomplish this, we obtained various types of seed text from several existing music tag and caption datasets and a knowledge graph dataset of artists and tracks. The experimental results show the effectiveness of TTMR++ in comparison to state-of-the-art music-text joint embedding models through a comprehensive evaluation involving various musical text queries.
We show that autoregressive decoding of a transformer-based language model can realize universal computation, without external intervention or modification of the model's weights. Establishing this result requires understanding how a language model can process arbitrarily long inputs using a bounded context. For this purpose, we consider a generalization of autoregressive decoding where, given a long input, emitted tokens are appended to the end of the sequence as the context window advances. We first show that the resulting system corresponds to a classical model of computation, a Lag system, that has long been known to be computationally universal. By leveraging a new proof, we show that a universal Turing machine can be simulated by a Lag system with 2027 production rules. We then investigate whether an existing large language model can simulate the behaviour of such a universal Lag system. We give an affirmative answer by showing that a single system-prompt can be developed for gemini-1.5-pro-001 that drives the model, under deterministic (greedy) decoding, to correctly apply each of the 2027 production rules. We conclude that, by the Church-Turing thesis, prompted gemini-1.5-pro-001 with extended autoregressive (greedy) decoding is a general purpose computer.
As large language models (LLMs) increasingly integrate into every aspect of our work and daily lives, there are growing concerns about user privacy, which push the trend toward local deployment of these models. There are a number of lightweight LLMs (e.g., Gemini Nano, LLAMA2 7B) that can run locally on smartphones, providing users with greater control over their personal data. As a rapidly emerging application, we are concerned about their performance on commercial-off-the-shelf mobile devices. To fully understand the current landscape of LLM deployment on mobile platforms, we conduct a comprehensive measurement study on mobile devices. We evaluate both metrics that affect user experience, including token throughput, latency, and battery consumption, as well as factors critical to developers, such as resource utilization, DVFS strategies, and inference engines. In addition, we provide a detailed analysis of how these hardware capabilities and system dynamics affect on-device LLM performance, which may help developers identify and address bottlenecks for mobile LLM applications. We also provide comprehensive comparisons across the mobile system-on-chips (SoCs) from major vendors, highlighting their performance differences in handling LLM workloads. We hope that this study can provide insights for both the development of on-device LLMs and the design for future mobile system architecture.
While retrieval augmented generation (RAG) has been shown to enhance factuality of large language model (LLM) outputs, LLMs still suffer from hallucination, generating incorrect or irrelevant information. One common detection strategy involves prompting the LLM again to assess whether its response is grounded in the retrieved evidence, but this approach is costly. Alternatively, lightweight natural language inference (NLI) models for efficient grounding verification can be used at inference time. While existing pre-trained NLI models offer potential solutions, their performance remains subpar compared to larger models on realistic RAG inputs. RAG inputs are more complex than most datasets used for training NLI models and have characteristics specific to the underlying knowledge base, requiring adaptation of the NLI models to a specific target domain. Additionally, the lack of labeled instances in the target domain makes supervised domain adaptation, e.g., through fine-tuning, infeasible. To address these challenges, we introduce Automatic Generative Domain Adaptation (Auto-GDA). Our framework enables unsupervised domain adaptation through synthetic data generation. Unlike previous methods that rely on handcrafted filtering and augmentation strategies, Auto-GDA employs an iterative process to continuously improve the quality of generated samples using weak labels from less efficient teacher models and discrete optimization to select the most promising augmented samples. Experimental results demonstrate the effectiveness of our approach, with models fine-tuned on synthetic data using Auto-GDA often surpassing the performance of the teacher model and reaching the performance level of LLMs at 10 % of their computational cost.
Deploying large language models (LLMs) is challenging due to their high memory and computational demands, especially during long-context inference. While key-value (KV) caching accelerates inference by reusing previously computed keys and values, it also introduces significant memory overhead. Existing KV cache compression methods such as eviction and merging typically compress the KV cache after it is generated and overlook the eviction of hidden states, failing to improve the speed of the prefilling stage. Additionally, applying a uniform compression rate across different attention heads can harm crucial retrieval heads in needle-in-a-haystack tasks due to excessive compression. In this paper, we propose UNComp, an uncertainty-aware compression scheme that leverages matrix entropy to estimate model uncertainty across layers and heads at the token sequence level. By grouping layers and heads based on their uncertainty, UNComp adaptively compresses both the hidden states and the KV cache. Our method achieves a 1.6x speedup in the prefilling stage and reduces the KV cache to 4.74% of its original size, resulting in a 6.4x increase in throughput and a 1.4x speedup in inference with only a 1.41% performance loss. Remarkably, in needle-in-a-haystack tasks, UNComp outperforms the full-size KV cache even when compressed to 9.38% of its original size. Our approach offers an efficient, training-free Grouped-Query Attention paradigm that can be seamlessly integrated into existing KV cache schemes.
We introduce Audio-Agent, a multimodal framework for audio generation, editing and composition based on text or video inputs. Conventional approaches for text-to-audio (TTA) tasks often make single-pass inferences from text descriptions. While straightforward, this design struggles to produce high-quality audio when given complex text conditions. In our method, we utilize a pre-trained TTA diffusion network as the audio generation agent to work in tandem with GPT-4, which decomposes the text condition into atomic, specific instructions and calls the agent for audio generation. In doing so, Audio-Agent can generate high-quality audio that is closely aligned with the provided text or video exhibiting complex and multiple events, while supporting variable-length and variable-volume generation. For video-to-audio (VTA) tasks, most existing methods require training a timestamp detector to synchronize video events with the generated audio, a process that can be tedious and time-consuming. Instead, we propose a simpler approach by fine-tuning a pre-trained Large Language Model (LLM), e.g., Gemma2-2B-it, to obtain both semantic and temporal conditions that bridge the video and audio modality. Consequently, our framework contributes a comprehensive solution for both TTA and VTA tasks without substantial computational overhead in training.
This paper examines how the sequencing of images and text within multi-modal prompts influences the reasoning performance of large language models (LLMs). We performed empirical evaluations using three commercial LLMs. Our results demonstrate that the order in which modalities are presented can significantly affect performance, particularly in tasks of varying complexity. For simpler tasks involving a single image, modality sequencing had a clear impact on accuracy. However, in more complex tasks involving multiple images and intricate reasoning steps, the effect of sequencing diminished, likely due to the increased cognitive demands of the task. Our findings also highlight the importance of question/prompt structure. In nested and multi-step reasoning tasks, modality sequencing played a key role in shaping model performance. While LLMs excelled in the initial stages of reasoning, they struggled to re-incorporate earlier information, underscoring the challenges of multi-hop reasoning within transformer architectures. This suggests that aligning the sequence of modalities with the logical flow of reasoning steps is more critical than modality order alone. These insights offer valuable implications for improving multi-modal prompt design, with broader applications across fields such as education, medical imaging, and cross-modal learning.
Throughout its lifecycle, a large language model (LLM) generates a substantially larger carbon footprint during inference than training. LLM inference requests vary in batch size, prompt length, and token generation number, while cloud providers employ different GPU types and quantities to meet diverse service-level objectives for accuracy and latency. It is crucial for both users and cloud providers to have a tool that quickly and accurately estimates the carbon impact of LLM inferences based on a combination of inference request and hardware configurations before execution. Estimating the carbon footprint of LLM inferences is more complex than training due to lower and highly variable model FLOPS utilization, rendering previous equation-based models inaccurate. Additionally, existing machine learning (ML) prediction methods either lack accuracy or demand extensive training data, as they inadequately handle the distinct prefill and decode phases, overlook hardware-specific features, and inefficiently sample uncommon inference configurations. We introduce \coo, a graph neural network (GNN)-based model that greatly improves the accuracy of LLM inference carbon footprint predictions compared to previous methods.
Keyphrase generation (KPG) aims to automatically generate a collection of phrases representing the core concepts of a given document. The dominant paradigms in KPG include one2seq and one2set. Recently, there has been increasing interest in applying large language models (LLMs) to KPG. Our preliminary experiments reveal that it is challenging for a single model to excel in both recall and precision. Further analysis shows that: 1) the one2set paradigm owns the advantage of high recall, but suffers from improper assignments of supervision signals during training; 2) LLMs are powerful in keyphrase selection, but existing selection methods often make redundant selections. Given these observations, we introduce a generate-then-select framework decomposing KPG into two steps, where we adopt a one2set-based model as generator to produce candidates and then use an LLM as selector to select keyphrases from these candidates. Particularly, we make two important improvements on our generator and selector: 1) we design an Optimal Transport-based assignment strategy to address the above improper assignments; 2) we model the keyphrase selection as a sequence labeling task to alleviate redundant selections. Experimental results on multiple benchmark datasets show that our framework significantly surpasses state-of-the-art models, especially in absent keyphrase prediction.