Large language models (LLM) have recently attracted surging interest due to their outstanding capabilities across various domains. However, enabling efficient LLM inference is challenging due to its autoregressive decoding that generates tokens only one at a time. Although research works apply pruning or quantization to speed up LLM inference, they typically require fine-tuning the LLM, incurring significant time and economic costs. Meanwhile, speculative decoding has been proposed to use small speculative models (SSMs) to accelerate the inference of LLM. However, the low acceptance rate of SSM and the high verification cost of LLM prohibit further performance improvement of inference. In this paper, we propose Minions, an LLM inference system that accelerates LLM inference with a collective and adaptive speculative generation. Specifically, Minions proposes a majority-voted mechanism to leverage multiple SSMs to jointly speculate the outputs of LLM, which improves the inference performance without introducing prohibitive computation costs for LLM. To better trade off the number of tokens speculated from SSM and the verification cost of LLM, Minions proposes an adaptive mechanism to dynamically determine the optimal speculation length of SSM, which can achieve better inference performance across different models, datasets, and hyper-parameters. In addition, Minions decouples the SSM decoding and LLM verification efficiently and adopts a pipelined execution mechanism to further improve the inference performance of LLM. By comparing with the state-of-the-art LLM inference systems, we demonstrate that Minions can achieve higher inference throughput and lower inference time.
相關內容
Evaluating the reasoning abilities of large language models (LLMs) is challenging. Existing benchmarks often depend on static datasets, which are vulnerable to data contamination and may get saturated over time, or on binary live human feedback that conflates reasoning with other abilities. As the most prominent dynamic benchmark, Chatbot Arena evaluates open-ended questions in real-world settings, but lacks the granularity in assessing specific reasoning capabilities. We introduce GameArena, a dynamic benchmark designed to evaluate LLM reasoning capabilities through interactive gameplay with humans. GameArena consists of three games designed to test specific reasoning capabilities (e.g., deductive and inductive reasoning), while keeping participants entertained and engaged. We analyze the gaming data retrospectively to uncover the underlying reasoning processes of LLMs and measure their fine-grained reasoning capabilities. We collect over 2000 game sessions and provide detailed assessments of various reasoning capabilities for five state-of-the-art LLMs. Our user study with 100 participants suggests that GameArena improves user engagement compared to Chatbot Arena. For the first time, GameArena enables the collection of step-by-step LLM reasoning data in the wild.
As large language models (LLMs) advance in their linguistic capacity, understanding how they capture aspects of language competence remains a significant challenge. This study therefore employs psycholinguistic paradigms in English, which are well-suited for probing deeper cognitive aspects of language processing, to explore neuron-level representations in language model across three tasks: sound-shape association, sound-gender association, and implicit causality. Our findings indicate that while GPT-2-XL struggles with the sound-shape task, it demonstrates human-like abilities in both sound-gender association and implicit causality. Targeted neuron ablation and activation manipulation reveal a crucial relationship: When GPT-2-XL displays a linguistic ability, specific neurons correspond to that competence; conversely, the absence of such an ability indicates a lack of specialized neurons. This study is the first to utilize psycholinguistic experiments to investigate deep language competence at the neuron level, providing a new level of granularity in model interpretability and insights into the internal mechanisms driving language ability in the transformer-based LLM.
Spatial perception is a fundamental component of intelligence. While many studies highlight that large multimodal language models (MLMs) struggle to reason about space, they only test for static spatial reasoning, such as categorizing the relative positions of objects. Meanwhile, real-world deployment requires dynamic capabilities like perspective-taking and egocentric action recognition. As a roadmap to improving spatial intelligence, we introduce SAT, Spatial Aptitude Training, which goes beyond static relative object position questions to the more dynamic tasks. SAT contains 218K question-answer pairs for 22K synthetic scenes across a training and testing set. Generated using a photo-realistic physics engine, our dataset can be arbitrarily scaled and easily extended to new actions, scenes, and 3D assets. We find that even MLMs that perform relatively well on static questions struggle to accurately answer dynamic spatial questions. Further, we show that SAT instruction-tuning data improves not only dynamic spatial reasoning on SAT, but also zero-shot performance on existing real-image spatial benchmarks: $23\%$ on CVBench, $8\%$ on the harder BLINK benchmark, and $18\%$ on VSR. When instruction-tuned on SAT, our 13B model matches larger proprietary MLMs like GPT4-V and Gemini-3-1.0 in spatial reasoning. Our data/code is available at //arijitray1993.github.io/SAT/ .
Automatic evaluation by large language models (LLMs) is a prominent topic today; however, judgment and evaluation tasks are often subjective and influenced by various factors, making adaptation challenging. While many studies demonstrate the capabilities of state-of-the-art proprietary LLMs in comparison to human evaluators, they often struggle to adapt to reference evaluators over time, a requirement for achieving personalized judgment. Additionally, numerous works have attempted to apply open LLMs as judges or evaluators, but these efforts frequently overlook the limitations of working with scarce data. Personalized judgment is inherently associated with limited data scenarios, which are common in many real-world problems. Our work aims to present a data augmentation technique to select a more effective sample from limited data in order to align an open LLM with human preference. Our work achieves approximately 7% improvements in Pearson correlation with a reference judge over the baseline,and 30% improvement over the base model (Llama3.1-8B-Instruct) in the mathematical reasoning evaluation task. demonstrating that augmenting selecting more effective preference data enables our approach to surpass baseline methods.
Ensembles of generative large language models (LLMs) can integrate the strengths of different LLMs to compensate for the limitations of individual models. However, recent work has focused on training an additional fusion model to combine complete responses from multiple LLMs, failing to tap into their collaborative potential to generate higher-quality responses. Moreover, as the additional fusion model is trained on a specialized dataset, these methods struggle with generalizing to open-domain queries from online users. In this paper, we propose SpecFuse, a novel ensemble framework that outputs the fused result by iteratively producing the next segment through collaboration among LLMs. This is achieved through cyclic execution of its inference and verification components. In each round, the inference component invokes each base LLM to generate candidate segments in parallel, and the verify component calls these LLMs again to predict the ranking of the segments. The top-ranked segment is then broadcast to all LLMs, encouraging them to generate higher-quality segments in the next round. This approach also allows the base LLMs to be plug-and-play, without any training or adaptation, avoiding generalization limitations. Furthermore, to conserve computational resources, we propose a model exit mechanism that dynamically excludes models exhibiting poor performance in previous rounds during each query response. In this way, it effectively reduces the number of model calls while maintaining overall performance.
LLM-BIP: Structured Pruning for Large Language Models with Block-Wise Forward Importance Propagation
Large language models (LLMs) have demonstrated remarkable performance across various language tasks, but their widespread deployment is impeded by their large size and high computational costs. Structural pruning is a prevailing technique used to introduce sparsity into pre-trained models and facilitate direct hardware acceleration during inference by removing redundant connections (structurally-grouped parameters), such as channels and attention heads. Existing structural pruning approaches often employ either global or layer-wise pruning criteria; however, they are hindered by ineffectiveness stemming from inaccurate evaluation of connection importance. Global pruning methods typically assess component importance using near-zero and unreliable gradients, while layer-wise pruning approaches encounter significant pruning error accumulation issues. To this end, we propose a more accurate pruning metric based on the block-wise importance score propagation, termed LLM-BIP. Specifically, LLM-BIP precisely evaluates connection importance by gauging its influence on the respective transformer block output, which can be efficiently approximated in a single forward pass through an upper bound derived from the assumption of Lipschitz continuity. We evaluate the proposed method using LLaMA-7B, Vicuna-7B, and LLaMA-13B across common zero-shot tasks. The results demonstrate that our approach achieves an average of 3.26% increase in accuracy for common reasoning tasks compared to previous best baselines. It also reduces perplexity by 14.09 and 68.76 on average for the WikiText2 dataset and PTB dataset, respectively.
Multimodal large language models (MLLMs) have achieved remarkable progress on various visual question answering and reasoning tasks leveraging instruction fine-tuning specific datasets. They can also learn from preference data annotated by human to enhance their reasoning ability and mitigate hallucinations. Most of preference data is generated from the model itself. However, existing methods require high-quality critical labels, which are costly and rely on human or proprietary models like GPT-4V. In this work, we propose Enhancing Alignment in MLLMs via Critical Observation (EACO), which aligns MLLMs by self-generated preference data using only 5k images economically. Our approach begins with collecting and refining a Scoring Evaluation Instruction-tuning dataset to train a critical evaluation model, termed the Critic. This Critic observes model responses across multiple dimensions, selecting preferred and non-preferred outputs for refined Direct Preference Optimization (DPO) tuning. To further enhance model performance, we employ an additional supervised fine-tuning stage after preference tuning. EACO reduces the overall hallucinations by 65.6% on HallusionBench and improves the reasoning ability by 21.8% on MME-Cognition. EACO achieves an 8.5% improvement over LLaVA-v1.6-Mistral-7B across multiple benchmarks. Remarkably, EACO also shows the potential critical ability in open-source MLLMs, demonstrating that EACO is a viable path to boost the competence of MLLMs.
Autoregressive language models demonstrate excellent performance in various scenarios. However, the inference efficiency is limited by its one-step-one-word generation mode, which has become a pressing problem recently as the models become increasingly larger. Speculative decoding employs a "draft and then verify" mechanism to allow multiple tokens to be generated in one step, realizing lossless acceleration. Existing methods mainly adopt fixed heuristic draft structures, which fail to adapt to different situations to maximize the acceptance length during verification. To alleviate this dilemma, we proposed OPT-Tree, an algorithm to construct adaptive and scalable draft trees. It searches the optimal tree structure that maximizes the mathematical expectation of the acceptance length in each decoding step. Experimental results reveal that OPT-Tree outperforms the existing draft structures and achieves a speed-up ratio of up to 3.2 compared with autoregressive decoding. If the draft model is powerful enough and the node budget is sufficient, it can generate more than ten tokens in a single step. Our code is available at //github.com/Jikai0Wang/OPT-Tree.
Large language models (LLMs) have demonstrated impressive capabilities in natural language processing. However, their internal mechanisms are still unclear and this lack of transparency poses unwanted risks for downstream applications. Therefore, understanding and explaining these models is crucial for elucidating their behaviors, limitations, and social impacts. In this paper, we introduce a taxonomy of explainability techniques and provide a structured overview of methods for explaining Transformer-based language models. We categorize techniques based on the training paradigms of LLMs: traditional fine-tuning-based paradigm and prompting-based paradigm. For each paradigm, we summarize the goals and dominant approaches for generating local explanations of individual predictions and global explanations of overall model knowledge. We also discuss metrics for evaluating generated explanations, and discuss how explanations can be leveraged to debug models and improve performance. Lastly, we examine key challenges and emerging opportunities for explanation techniques in the era of LLMs in comparison to conventional machine learning models.
The emergence of large language models (LLMs) has substantially influenced natural language processing, demonstrating exceptional results across various tasks. In this study, we employ ``Introspective Tips" to facilitate LLMs in self-optimizing their decision-making. By introspectively examining trajectories, LLM refines its policy by generating succinct and valuable tips. Our method enhances the agent's performance in both few-shot and zero-shot learning situations by considering three essential scenarios: learning from the agent's past experiences, integrating expert demonstrations, and generalizing across diverse games. Importantly, we accomplish these improvements without fine-tuning the LLM parameters; rather, we adjust the prompt to generalize insights from the three aforementioned situations. Our framework not only supports but also emphasizes the advantage of employing LLM in in-contxt decision-making. Experiments involving over 100 games in TextWorld illustrate the superior performance of our approach.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191