The increased use of large language models (LLMs) across a variety of real-world applications calls for automatic tools to check the factual accuracy of their outputs, as LLMs often hallucinate. This is difficult as it requires assessing the factuality of free-form open-domain responses. While there has been a lot of research on this topic, different papers use different evaluation benchmarks and measures, which makes them hard to compare and hampers future progress. To mitigate these issues, we developed OpenFactCheck, a unified framework, with three modules: (i) RESPONSEEVAL, which allows users to easily customize an automatic fact-checking system and to assess the factuality of all claims in an input document using that system, (ii) LLMEVAL, which assesses the overall factuality of an LLM, and (iii) CHECKEREVAL, a module to evaluate automatic fact-checking systems. OpenFactCheck is open-sourced (//github.com/mbzuai-nlp/openfactcheck) and publicly released as a Python library (//pypi.org/project/openfactcheck/) and also as a web service (//app.openfactcheck.com). A video describing the system is available at //youtu.be/-i9VKL0HleI.
相關內容
Multimodal large language models (MLLMs) combine visual and textual data for tasks such as image captioning and visual question answering. Proper uncertainty calibration is crucial, yet challenging, for reliable use in areas like healthcare and autonomous driving. This paper investigates representative MLLMs, focusing on their calibration across various scenarios, including before and after visual fine-tuning, as well as before and after multimodal training of the base LLMs. We observed miscalibration in their performance, and at the same time, no significant differences in calibration across these scenarios. We also highlight how uncertainty differs between text and images and how their integration affects overall uncertainty. To better understand MLLMs' miscalibration and their ability to self-assess uncertainty, we construct the IDK (I don't know) dataset, which is key to evaluating how they handle unknowns. Our findings reveal that MLLMs tend to give answers rather than admit uncertainty, but this self-assessment improves with proper prompt adjustments. Finally, to calibrate MLLMs and enhance model reliability, we propose techniques such as temperature scaling and iterative prompt optimization. Our results provide insights into improving MLLMs for effective and responsible deployment in multimodal applications. Code and IDK dataset: \href{//github.com/hfutml/Calibration-MLLM}{//github.com/hfutml/Calibration-MLLM}.
Recent generative large language models (LLMs) show remarkable performance in non-English languages, but when prompted in those languages they tend to express higher harmful social biases and toxicity levels. Prior work has shown that finetuning on specialized datasets can mitigate this behavior, and doing so in English can transfer to other languages. In this work, we investigate the impact of different finetuning methods on the model's bias and toxicity, but also on its ability to produce fluent and diverse text. Our results show that finetuning on curated non-harmful text is more effective for mitigating bias, and finetuning on direct preference optimization (DPO) datasets is more effective for mitigating toxicity. The mitigation caused by applying these methods in English also transfers to non-English languages. We find evidence that the extent to which transfer takes place can be predicted by the amount of data in a given language present in the model's pretraining data. However, this transfer of bias and toxicity mitigation often comes at the expense of decreased language generation ability in non-English languages, highlighting the importance of developing language-specific bias and toxicity mitigation methods.
Convolutional Neural Networks (CNNs) are crucial in various applications, but their deployment on resource-constrained edge devices poses challenges. This study presents the Sum-of-Products (SOP) units for convolution, which utilize low-latency left-to-right bit-serial arithmetic to minimize response time and enhance overall performance. The study proposes a methodology for fusing multiple convolution layers to reduce off-chip memory communication and increase overall performance. An effective mechanism detects and skips inefficient convolutions after ReLU layers, minimizing power consumption without compromising accuracy. Furthermore, efficient tile movement guarantees uniform access to the fusion pyramid. An analysis demonstrates the utile stride strategy improves operational intensity. Two designs cater to varied demands: one focuses on minimal response time for mission-critical applications, and another focuses on resource-constrained devices with comparable latency. This approach notably reduced redundant computations, improving the efficiency of CNN deployment on edge devices.
The exceptional capabilities of large language models (LLMs) have substantially accelerated the rapid rise and widespread adoption of agents. Recent studies have demonstrated that generating Python code to consolidate LLM-based agents' actions into a unified action space (CodeAct) is a promising approach for developing real-world LLM agents. However, this step-by-step code generation approach often lacks consistency and robustness, leading to instability in agent applications, particularly for complex reasoning and out-of-domain tasks. In this paper, we propose a novel approach called Tree-of-Code (ToC) to tackle the challenges of complex problem planning and execution with an end-to-end mechanism. By integrating key ideas from both Tree-of-Thought and CodeAct, ToC combines their strengths to enhance solution exploration. In our framework, each final code execution result is treated as a node in the decision tree, with a breadth-first search strategy employed to explore potential solutions. The final outcome is determined through a voting mechanism based on the outputs of the nodes.
Large Language Models (LLMs) are increasingly deployed in real-world applications that demand complex reasoning. To track progress, robust benchmarks are required to evaluate their capabilities beyond superficial pattern recognition. However, current LLM reasoning benchmarks often face challenges such as insufficient interpretability, performance saturation or data contamination. To address these challenges, we introduce GAMEBoT, a gaming arena designed for rigorous and transparent assessment of LLM reasoning capabilities. GAMEBoT decomposes complex reasoning in games into predefined modular subproblems. This decomposition allows us to design a suite of Chain-of-Thought (CoT) prompts that leverage domain knowledge to guide LLMs in addressing these subproblems before action selection. Furthermore, we develop a suite of rule-based algorithms to generate ground truth for these subproblems, enabling rigorous validation of the LLMs' intermediate reasoning steps. This approach facilitates evaluation of both the quality of final actions and the accuracy of the underlying reasoning process. GAMEBoT also naturally alleviates the risk of data contamination through dynamic games and head-to-head LLM competitions. We benchmark 17 prominent LLMs across eight games, encompassing various strategic abilities and game characteristics. Our results suggest that GAMEBoT presents a significant challenge, even when LLMs are provided with detailed CoT prompts. Project page: \url{//visual-ai.github.io/gamebot}
With the integration of large language models (LLMs), embodied agents have strong capabilities to execute complicated instructions in natural language, paving a way for the potential deployment of embodied robots. However, a foreseeable issue is that those embodied agents can also flawlessly execute some hazardous tasks, potentially causing damages in real world. To study this issue, we present SafeAgentBench -- a new benchmark for safety-aware task planning of embodied LLM agents. SafeAgentBench includes: (1) a new dataset with 750 tasks, covering 10 potential hazards and 3 task types; (2) SafeAgentEnv, a universal embodied environment with a low-level controller, supporting multi-agent execution with 17 high-level actions for 8 state-of-the-art baselines; and (3) reliable evaluation methods from both execution and semantic perspectives. Experimental results show that the best-performing baseline gets 69% success rate for safe tasks, but only 5% rejection rate for hazardous tasks, indicating significant safety risks. More details and codes are available at //github.com/shengyin1224/SafeAgentBench.
Selective classification frameworks are useful tools for automated decision making in highly risky scenarios, since they allow for a classifier to only make highly confident decisions, while abstaining from making a decision when it is not confident enough to do so, which is otherwise known as an indecision. For a given level of classification accuracy, we aim to make as many decisions as possible. For many problems, this can be achieved without abstaining from making decisions. But when the problem is hard enough, we show that we can still control the misclassification rate of a classifier up to any user specified level, while only abstaining from the minimum necessary amount of decisions, even if this level of misclassification is smaller than the Bayes optimal error rate. In many problem settings, the user could obtain a dramatic decrease in misclassification while only paying a comparatively small price in terms of indecisions.
Visual Language Models (VLMs) demonstrate impressive capabilities in processing multimodal inputs, yet applications such as visual agents, which require handling multiple images and high-resolution videos, demand enhanced long-range modeling. Moreover, existing open-source VLMs lack systematic exploration into extending their context length, and commercial models often provide limited details. To tackle this, we aim to establish an effective solution that enhances long context performance of VLMs while preserving their capacities in short context scenarios. Towards this goal, we make the best design choice through extensive experiment settings from data curation to context window extending and utilizing: (1) we analyze data sources and length distributions to construct ETVLM - a data recipe to balance the performance across scenarios; (2) we examine existing position extending methods, identify their limitations and propose M-RoPE++ as an enhanced approach; we also choose to solely instruction-tune the backbone with mixed-source data; (3) we discuss how to better utilize extended context windows and propose hybrid-resolution training. Built on the Qwen-VL series model, we propose Giraffe, which is effectively extended to 128K lengths. Evaluated on extensive long context VLM benchmarks such as VideoMME and Viusal Haystacks, our Giraffe achieves state-of-the-art performance among similarly sized open-source long VLMs and is competitive with commercial model GPT-4V. We will open-source the code, data, and models.
The recent success of large language models (LLMs) trained on static, pre-collected, general datasets has sparked numerous research directions and applications. One such direction addresses the non-trivial challenge of integrating pre-trained LLMs into dynamic data distributions, task structures, and user preferences. Pre-trained LLMs, when tailored for specific needs, often experience significant performance degradation in previous knowledge domains -- a phenomenon known as "catastrophic forgetting". While extensively studied in the continual learning (CL) community, it presents new manifestations in the realm of LLMs. In this survey, we provide a comprehensive overview of the current research progress on LLMs within the context of CL. This survey is structured into four main sections: we first describe an overview of continually learning LLMs, consisting of two directions of continuity: vertical continuity (or vertical continual learning), i.e., continual adaptation from general to specific capabilities, and horizontal continuity (or horizontal continual learning), i.e., continual adaptation across time and domains (Section 3). We then summarize three stages of learning LLMs in the context of modern CL: Continual Pre-Training (CPT), Domain-Adaptive Pre-training (DAP), and Continual Fine-Tuning (CFT) (Section 4). Then we provide an overview of evaluation protocols for continual learning with LLMs, along with the current available data sources (Section 5). Finally, we discuss intriguing questions pertaining to continual learning for LLMs (Section 6). The full list of papers examined in this survey is available at //github.com/Wang-ML-Lab/llm-continual-learning-survey.
While large language models (LLMs) have demonstrated remarkable capabilities across a range of downstream tasks, a significant concern revolves around their propensity to exhibit hallucinations: LLMs occasionally generate content that diverges from the user input, contradicts previously generated context, or misaligns with established world knowledge. This phenomenon poses a substantial challenge to the reliability of LLMs in real-world scenarios. In this paper, we survey recent efforts on the detection, explanation, and mitigation of hallucination, with an emphasis on the unique challenges posed by LLMs. We present taxonomies of the LLM hallucination phenomena and evaluation benchmarks, analyze existing approaches aiming at mitigating LLM hallucination, and discuss potential directions for future research.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191