Performance evaluation plays a crucial role in the development life cycle of large language models (LLMs). It estimates the model's capability, elucidates behavior characteristics, and facilitates the identification of potential issues and limitations, thereby guiding further improvement. Given that LLMs' diverse task-handling abilities stem from large volumes of training data, a comprehensive evaluation also necessitates abundant, well-annotated, and representative test data to assess LLM performance across various downstream tasks. However, the demand for high-quality test data often entails substantial time, computational resources, and manual efforts, sometimes causing the evaluation to be inefficient or impractical. To address these challenges, researchers propose active testing, which estimates the overall performance by selecting a subset of test data. Nevertheless, the existing active testing methods tend to be inefficient, even inapplicable, given the unique new challenges of LLMs (e.g., diverse task types, increased model complexity, and unavailability of training data). To mitigate such limitations and expedite the development cycle of LLMs, in this work, we introduce AcTracer, an active testing framework tailored for LLMs that strategically selects a small subset of test data to achieve a more accurate performance estimation for LLMs. AcTracer utilizes both internal and external information from LLMs to guide the test sampling process, reducing variance through a multi-stage pool-based active selection. Our experiment results demonstrate that AcTracer achieves state-of-the-art performance compared to existing methods across various tasks.
We introduce EXAONE 3.0 instruction-tuned language model, the first open model in the family of Large Language Models (LLMs) developed by LG AI Research. Among different model sizes, we publicly release the 7.8B instruction-tuned model to promote open research and innovations. Through extensive evaluations across a wide range of public and in-house benchmarks, EXAONE 3.0 demonstrates highly competitive real-world performance with instruction-following capability against other state-of-the-art open models of similar size. Our comparative analysis shows that EXAONE 3.0 excels particularly in Korean, while achieving compelling performance across general tasks and complex reasoning. With its strong real-world effectiveness and bilingual proficiency, we hope that EXAONE keeps contributing to advancements in Expert AI. Our EXAONE 3.0 instruction-tuned model is available at https://huggingface.co/LGAI-EXAONE/EXAONE-3.0-7.8B-Instruct
In this paper we demonstrate how logic programming systems and Automated first-order logic Theorem Provers (ATPs) can improve the accuracy of Large Language Models (LLMs) for logical reasoning tasks where the baseline performance is given by direct LLM solutions. We first evaluate LLM reasoning on steamroller problems using the PRONTOQA benchmark. We show how accuracy can be improved with a neuro-symbolic architecture where the LLM acts solely as a front-end for translating a given problem into a formal logic language and an automated reasoning engine is called for solving it. However, this approach critically hinges on the correctness of the LLM translation. To assess this translation correctness, we secondly define a framework of syntactic and semantic error categories. We implemented the framework and used it to identify errors that LLMs make in the benchmark domain. Based on these findings, we thirdly extended our method with capabilities for automatically correcting syntactic and semantic errors. For semantic error correction we integrate first-order logic ATPs, which is our main and novel contribution. We demonstrate that this approach reduces semantic errors significantly and further increases the accurracy of LLM logical reasoning.
Binding affinity optimization is crucial in early-stage drug discovery. While numerous machine learning methods exist for predicting ligand potency, their comparative efficacy remains unclear. This study evaluates the performance of classical tree-based models and advanced neural networks in protein-ligand binding affinity prediction. Our comprehensive benchmarking encompasses 2D models utilizing ligand-only RDKit embeddings and Large Language Model (LLM) ligand representations, as well as 3D neural networks incorporating bound protein-ligand conformations. We assess these models across multiple standard datasets, examining various predictive scenarios including classification, ranking, regression, and active learning. Results indicate that simpler models can surpass more complex ones in specific tasks, while 3D models leveraging structural information become increasingly competitive with larger training datasets containing compounds with labelled affinity data against multiple targets. Pre-trained 3D models, by incorporating protein pocket environments, demonstrate significant advantages in data-scarce scenarios for specific binding pockets. Additionally, LLM pretraining on 2D ligand data enhances complex model performance, providing versatile embeddings that outperform traditional RDKit features in computational efficiency. Finally, we show that combining 2D and 3D model strengths improves active learning outcomes beyond current state-of-the-art approaches. These findings offer valuable insights for optimizing machine learning strategies in drug discovery pipelines.
StreamVoice has recently pushed the boundaries of zero-shot voice conversion (VC) in the streaming domain. It uses a streamable language model (LM) with a context-aware approach to convert semantic features from automatic speech recognition (ASR) into acoustic features with the desired speaker timbre. Despite its innovations, StreamVoice faces challenges due to its dependency on a streaming ASR within a cascaded framework, which complicates system deployment and optimization, affects VC system's design and performance based on the choice of ASR, and struggles with conversion stability when faced with low-quality semantic inputs. To overcome these limitations, we introduce StreamVoice+, an enhanced LM-based end-to-end streaming framework that operates independently of streaming ASR. StreamVoice+ integrates a semantic encoder and a connector with the original StreamVoice framework, now trained using a non-streaming ASR. This model undergoes a two-stage training process: initially, the StreamVoice backbone is pre-trained for voice conversion and the semantic encoder for robust semantic extraction. Subsequently, the system is fine-tuned end-to-end, incorporating a LoRA matrix to activate comprehensive streaming functionality. Furthermore, StreamVoice+ mainly introduces two strategic enhancements to boost conversion quality: a residual compensation mechanism in the connector to ensure effective semantic transmission and a self-refinement strategy that leverages pseudo-parallel speech pairs generated by the conversion backbone to improve speech decoupling. Experiments demonstrate that StreamVoice+ not only achieves higher naturalness and speaker similarity in voice conversion than its predecessor but also provides versatile support for both streaming and non-streaming conversion scenarios.
Large language model (LLM) has recently been considered a promising technique for many fields. This work explores LLM-based wireless network optimization via in-context learning. To showcase the potential of LLM technologies, we consider the base station (BS) power control as a case study, a fundamental but crucial technique that is widely investigated in wireless networks. Different from existing machine learning (ML) methods, our proposed in-context learning algorithm relies on LLM's inference capabilities. It avoids the complexity of tedious model training and hyper-parameter fine-tuning, which is a well-known bottleneck of many ML algorithms. Specifically, the proposed algorithm first describes the target task via formatted natural language, and then designs the in-context learning framework and demonstration examples. After that, it considers two cases, namely discrete-state and continuous-state problems, and proposes state-based and ranking-based methods to select appropriate examples for these two cases, respectively. Finally, the simulations demonstrate that the proposed algorithm can achieve comparable performance as conventional deep reinforcement learning (DRL) techniques without dedicated model training or fine-tuning. Such an efficient and low-complexity approach has great potential for future wireless network optimization.
As content generated by Large Language Model (LLM) has grown exponentially, the ability to accurately identify and fingerprint such text has become increasingly crucial. In this work, we introduce a novel black-box approach for fingerprinting LLMs, achieving an impressive 72% accuracy in identifying the correct family of models (Such as Llama, Mistral, Gemma, etc) among a lineup of LLMs. We present an evolutionary strategy that leverages the capabilities of one LLM to discover the most salient features for identifying other LLMs. Our method employs a unique "Hide and Seek" algorithm, where an Auditor LLM generates discriminative prompts, and a Detective LLM analyzes the responses to fingerprint the target models. This approach not only demonstrates the feasibility of LLM-driven model identification but also reveals insights into the semantic manifolds of different LLM families. By iteratively refining prompts through in-context learning, our system uncovers subtle distinctions between model outputs, providing a powerful tool for LLM analysis and verification. This research opens new avenues for understanding LLM behavior and has significant implications for model attribution, security, and the broader field of AI transparency.
This paper presents and evaluates a new retrieval augmented generation (RAG) and large language model (LLM)-based artificial intelligence (AI) technique: rubric enabled generative artificial intelligence (REGAI). REGAI uses rubrics, which can be created manually or automatically by the system, to enhance the performance of LLMs for evaluation purposes. REGAI improves on the performance of both classical LLMs and RAG-based LLM techniques. This paper describes REGAI, presents data regarding its performance and discusses several possible application areas for the technology.
Embodied AI represents systems where AI is integrated into physical entities. Large Language Model (LLM), which exhibits powerful language understanding abilities, has been extensively employed in embodied AI by facilitating sophisticated task planning. However, a critical safety issue remains overlooked: could these embodied LLMs perpetrate harmful behaviors? In response, we introduce BadRobot, a novel attack paradigm aiming to make embodied LLMs violate safety and ethical constraints through typical voice-based user-system interactions. Specifically, three vulnerabilities are exploited to achieve this type of attack: (i) manipulation of LLMs within robotic systems, (ii) misalignment between linguistic outputs and physical actions, and (iii) unintentional hazardous behaviors caused by world knowledge's flaws. Furthermore, we construct a benchmark of various malicious physical action queries to evaluate BadRobot's attack performance. Based on this benchmark, extensive experiments against existing prominent embodied LLM frameworks (e.g., Voxposer, Code as Policies, and ProgPrompt) demonstrate the effectiveness of our BadRobot.
Interactive virtual humanoid agent is a crucial interface with the physical world. A relatively complete humanoid agent first needs to have face and body, then possess both verbal and non-verbal (such as eye contact, facial expression, lip motion, gesture, and manipulation) abilities, and finally, it is capable of real-time duplex communication, e.g., the ability to actively interrupt conversations. Most prior systems typically only consider a subset of these elements, leaving a gap from realistic humanoid agent. In this work, we propose a real-time, duplex, interactive end-to-end network capable of modeling realistic agent behaviors, including speech, full-body movements for talking, responding, idling, and manipulation. This system is a multimodal model integrating audio and visual inputs, extended from a pre-trained large language model (LLM). We collect approximately 200,000 hours of audio, around 130,000 hours of video data, and about 20,000 alignment samples to build the model. The final model demonstrates capabilities that are difficult to achieve in previous systems, such as generalized object manipulation. This work performs a preliminary exploration of the end-to-end approach in this field, aiming to inspire further research towards scaling up.
Evaluation is the baton for the development of large language models. Current evaluations typically employ a single-item assessment paradigm for each atomic test objective, which struggles to discern whether a model genuinely possesses the required capabilities or merely memorizes/guesses the answers to specific questions. To this end, we propose a novel evaluation framework referred to as StructEval. Starting from an atomic test objective, StructEval deepens and broadens the evaluation by conducting a structured assessment across multiple cognitive levels and critical concepts, and therefore offers a comprehensive, robust and consistent evaluation for LLMs. Experiments on three widely-used benchmarks demonstrate that StructEval serves as a reliable tool for resisting the risk of data contamination and reducing the interference of potential biases, thereby providing more reliable and consistent conclusions regarding model capabilities. Our framework also sheds light on the design of future principled and trustworthy LLM evaluation protocols.
The rapid increase in multimedia data has spurred advancements in Multimodal Summarization with Multimodal Output (MSMO), which aims to produce a multimodal summary that integrates both text and relevant images. The inherent heterogeneity of content within multimodal inputs and outputs presents a significant challenge to the execution of MSMO. Traditional approaches typically adopt a holistic perspective on coarse image-text data or individual visual objects, overlooking the essential connections between objects and the entities they represent. To integrate the fine-grained entity knowledge, we propose an Entity-Guided Multimodal Summarization model (EGMS). Our model, building on BART, utilizes dual multimodal encoders with shared weights to process text-image and entity-image information concurrently. A gating mechanism then combines visual data for enhanced textual summary generation, while image selection is refined through knowledge distillation from a pre-trained vision-language model. Extensive experiments on public MSMO dataset validate the superiority of the EGMS method, which also prove the necessity to incorporate entity information into MSMO problem.
This paper describes our approach to the SemEval-2024 safe biomedical Natural Language Inference for Clinical Trials (NLI4CT) task, which concerns classifying statements about Clinical Trial Reports (CTRs). We explored the capabilities of Mistral-7B, a generalist open-source Large Language Model (LLM). We developed a prompt for the NLI4CT task, and fine-tuned a quantized version of the model using an augmented version of the training dataset. The experimental results show that this approach can produce notable results in terms of the macro F1-score, while having limitations in terms of faithfulness and consistency. All the developed code is publicly available on a GitHub repository
Prompt compression is crucial for enhancing inference speed, reducing costs, and improving user experience. However, current methods face challenges such as low compression ratios and potential data leakage during evaluation. To address these issues, we propose 500xCompressor, a method that compresses extensive natural language contexts into a minimum of one single special token. The 500xCompressor introduces approximately 0.3% additional parameters and achieves compression ratios ranging from 6x to 480x. It is designed to compress any text, answer various types of questions, and could be utilized by the original large language model (LLM) without requiring fine-tuning. Initially, 500xCompressor was pretrained on the Arxiv Corpus, followed by fine-tuning on the ArxivQA dataset, and subsequently evaluated on strictly unseen and classical question answering (QA) datasets. The results demonstrate that the LLM retained 62.26-72.89% of its capabilities compared to using non-compressed prompts. This study also shows that not all the compressed tokens are equally utilized and that K V values have significant advantages over embeddings in preserving information at high compression ratios. The highly compressive nature of natural language prompts, even for fine-grained complex information, suggests promising potential for future applications and further research into developing a new LLM language.
Understanding how language and linguistic constructions are processed in the brain is a fundamental question in cognitive computational neuroscience. In this study, we explore the representation and processing of Argument Structure Constructions (ASCs) in a recurrent neural language model. We trained a Long Short-Term Memory (LSTM) network on a custom-made dataset consisting of 2000 sentences, generated using GPT-4, representing four distinct ASCs: transitive, ditransitive, caused-motion, and resultative constructions. We analyzed the internal activations of the LSTM model's hidden layers using Multidimensional Scaling (MDS) and t-Distributed Stochastic Neighbor Embedding (t-SNE) to visualize the sentence representations. The Generalized Discrimination Value (GDV) was calculated to quantify the degree of clustering within these representations. Our results show that sentence representations form distinct clusters corresponding to the four ASCs across all hidden layers, with the most pronounced clustering observed in the last hidden layer before the output layer. This indicates that even a relatively simple, brain-constrained recurrent neural network can effectively differentiate between various construction types. These findings are consistent with previous studies demonstrating the emergence of word class and syntax rule representations in recurrent language models trained on next word prediction tasks. In future work, we aim to validate these results using larger language models and compare them with neuroimaging data obtained during continuous speech perception. This study highlights the potential of recurrent neural language models to mirror linguistic processing in the human brain, providing valuable insights into the computational and neural mechanisms underlying language understanding.
Retrieving external knowledge and prompting large language models with relevant information is an effective paradigm to enhance the performance of question-answering tasks. Previous research typically handles paragraphs from external documents in isolation, resulting in a lack of context and ambiguous references, particularly in multi-document and complex tasks. To overcome these challenges, we propose a new retrieval framework IIER, that leverages Inter-chunk Interactions to Enhance Retrieval. This framework captures the internal connections between document chunks by considering three types of interactions: structural, keyword, and semantic. We then construct a unified Chunk-Interaction Graph to represent all external documents comprehensively. Additionally, we design a graph-based evidence chain retriever that utilizes previous paths and chunk interactions to guide the retrieval process. It identifies multiple seed nodes based on the target question and iteratively searches for relevant chunks to gather supporting evidence. This retrieval process refines the context and reasoning chain, aiding the large language model in reasoning and answer generation. Extensive experiments demonstrate that IIER outperforms strong baselines across four datasets, highlighting its effectiveness in improving retrieval and reasoning capabilities.
Current research on RAGs is distributed across various disciplines, and since the technology is evolving very quickly, its unit of analysis is mostly on technological innovations, rather than applications in business contexts. Thus, in this research, we aim to create a taxonomy to conceptualize a comprehensive overview of the constituting characteristics that define RAG applications, facilitating the adoption of this technology in the IS community. To the best of our knowledge, no RAG application taxonomies have been developed so far. We describe our methodology for developing the taxonomy, which includes the criteria for selecting papers, an explanation of our rationale for employing a Large Language Model (LLM)-supported approach to extract and identify initial characteristics, and a concise overview of our systematic process for conceptualizing the taxonomy. Our systematic taxonomy development process includes four iterative phases designed to refine and enhance our understanding and presentation of RAG's core dimensions. We have developed a total of five meta-dimensions and sixteen dimensions to comprehensively capture the concept of Retrieval-Augmented Generation (RAG) applications. When discussing our findings, we also detail the specific research areas and pose key research questions to guide future information system researchers as they explore the emerging topics of RAG systems.
Visual language navigation (VLN) is one of the important research in embodied AI. It aims to enable an agent to understand the surrounding environment and complete navigation tasks. VLN instructions could be categorized into coarse-grained and fine-grained commands. Fine-grained command describes a whole task with subtasks step-by-step. In contrast, coarse-grained command gives an abstract task description, which more suites human habits. Most existing work focuses on the former kind of instruction in VLN tasks, ignoring the latter abstract instructions belonging to daily life scenarios. To overcome the above challenge in abstract instruction, we attempt to consider coarse-grained instruction in VLN by event knowledge enhancement. Specifically, we first propose a prompt-based framework to extract an event knowledge graph (named VLN-EventKG) for VLN integrally over multiple mainstream benchmark datasets. Through small and large language model collaboration, we realize knowledge-enhanced navigation planning (named EventNav) for VLN tasks with coarse-grained instruction input. Additionally, we design a novel dynamic history backtracking module to correct potential error action planning in real time. Experimental results in various public benchmarks show our knowledge-enhanced method has superiority in coarse-grained-instruction VLN using our proposed VLN-EventKG with over $5\%$ improvement in success rate. Our project is available at https://sites.google.com/view/vln-eventkg
Evaluating the quality of automatically generated image descriptions is challenging, requiring metrics that capture various aspects such as grammaticality, coverage, correctness, and truthfulness. While human evaluation offers valuable insights, its cost and time-consuming nature pose limitations. Existing automated metrics like BLEU, ROUGE, METEOR, and CIDEr aim to bridge this gap but often show weak correlations with human judgment. We address this challenge by introducing a novel evaluation framework rooted in a modern large language model (LLM), such as GPT-4 or Gemini, capable of image generation. In our proposed framework, we begin by feeding an input image into a designated image captioning model, chosen for evaluation, to generate a textual description. Using this description, an LLM then creates a new image. By extracting features from both the original and LLM-created images, we measure their similarity using a designated similarity metric. A high similarity score suggests that the image captioning model has accurately generated textual descriptions, while a low similarity score indicates discrepancies, revealing potential shortcomings in the model's performance. Human-annotated reference captions are not required in our proposed evaluation framework, which serves as a valuable tool for evaluating the effectiveness of image captioning models. Its efficacy is confirmed through human evaluation.
To avoid detection, adversaries often use command-line obfuscation. There are numerous techniques of the command-line obfuscation, all designed to alter the command-line syntax without affecting its original functionality. This variability forces most security solutions to create an exhaustive enumeration of signatures for even a single pattern. In contrast to using signatures, we have implemented a scalable NLP-based detection method that leverages a custom-trained, small transformer language model that can be applied to any source of execution logs. The evaluation on top of real-world telemetry demonstrates that our approach yields high-precision detections even on high-volume telemetry from a diverse set of environments spanning from universities and businesses to healthcare or finance. The practical value is demonstrated in a case study of real-world samples detected by our model. We show the model's superiority to signatures on established malware known to employ obfuscation and showcase previously unseen obfuscated samples detected by our model.