Abstract:Alignment is a crucial step to enhance the instruction-following and conversational abilities of language models. Despite many recent work proposing new algorithms, datasets, and training pipelines, there is a lack of comprehensive studies measuring the impact of various design choices throughout the whole training process. We first conduct a rigorous analysis over a three-stage training pipeline consisting of supervised fine-tuning, offline preference learning, and online preference learning. We have found that using techniques like sequence packing, loss masking in SFT, increasing the preference dataset size in DPO, and online DPO training can significantly improve the performance of language models. We then train from Gemma-2b-base and LLama-3-8b-base, and find that our best models exceed the performance of the official instruct models tuned with closed-source data and algorithms. Our code and models can be found at https://github.com/Columbia-NLP-Lab/LionAlignment.
Abstract:We introduce DiffuseST, a low-latency, direct speech-to-speech translation system capable of preserving the input speaker's voice zero-shot while translating from multiple source languages into English. We experiment with the synthesizer component of the architecture, comparing a Tacotron-based synthesizer to a novel diffusion-based synthesizer. We find the diffusion-based synthesizer to improve MOS and PESQ audio quality metrics by 23\% each and speaker similarity by 5\% while maintaining comparable BLEU scores. Despite having more than double the parameter count, the diffusion synthesizer has lower latency, allowing the entire model to run more than 5$\times$ faster than real-time.
Abstract:Learning-based underwater image enhancement (UIE) methods have made great progress. However, the lack of large-scale and high-quality paired training samples has become the main bottleneck hindering the development of UIE. The inter-frame information in underwater videos can accelerate or optimize the UIE process. Thus, we constructed the first large-scale high-resolution underwater video enhancement benchmark (UVEB) to promote the development of underwater vision.It contains 1,308 pairs of video sequences and more than 453,000 high-resolution with 38\% Ultra-High-Definition (UHD) 4K frame pairs. UVEB comes from multiple countries, containing various scenes and video degradation types to adapt to diverse and complex underwater environments. We also propose the first supervised underwater video enhancement method, UVE-Net. UVE-Net converts the current frame information into convolutional kernels and passes them to adjacent frames for efficient inter-frame information exchange. By fully utilizing the redundant degraded information of underwater videos, UVE-Net completes video enhancement better. Experiments show the effective network design and good performance of UVE-Net.
Abstract:Retrieval-augmented question-answering systems combine retrieval techniques with large language models to provide answers that are more accurate and informative. Many existing toolkits allow users to quickly build such systems using off-the-shelf models, but they fall short in supporting researchers and developers to customize the model training, testing, and deployment process. We propose LocalRQA, an open-source toolkit that features a wide selection of model training algorithms, evaluation methods, and deployment tools curated from the latest research. As a showcase, we build QA systems using online documentation obtained from Databricks and Faire's websites. We find 7B-models trained and deployed using LocalRQA reach a similar performance compared to using OpenAI's text-ada-002 and GPT-4-turbo.
Abstract:In recent years, formal methods have been extensively used in the design of autonomous systems. By employing mathematically rigorous techniques, formal methods can provide fully automated reasoning processes with provable safety guarantees for complex dynamic systems with intricate interactions between continuous dynamics and discrete logics. This paper provides a comprehensive review of formal controller synthesis techniques for safety-critical autonomous systems. Specifically, we categorize the formal control synthesis problem based on diverse system models, encompassing deterministic, non-deterministic, and stochastic, and various formal safety-critical specifications involving logic, real-time, and real-valued domains. The review covers fundamental formal control synthesis techniques, including abstraction-based approaches and abstraction-free methods. We explore the integration of data-driven synthesis approaches in formal control synthesis. Furthermore, we review formal techniques tailored for multi-agent systems (MAS), with a specific focus on various approaches to address the scalability challenges in large-scale systems. Finally, we discuss some recent trends and highlight research challenges in this area.
Abstract:A reliable resume-job matching system helps a company find suitable candidates from a pool of resumes, and helps a job seeker find relevant jobs from a list of job posts. However, since job seekers apply only to a few jobs, interaction records in resume-job datasets are sparse. Different from many prior work that use complex modeling techniques, we tackle this sparsity problem using data augmentations and a simple contrastive learning approach. ConFit first creates an augmented resume-job dataset by paraphrasing specific sections in a resume or a job post. Then, ConFit uses contrastive learning to further increase training samples from $B$ pairs per batch to $O(B^2)$ per batch. We evaluate ConFit on two real-world datasets and find it outperforms prior methods (including BM25 and OpenAI text-ada-002) by up to 19% and 31% absolute in nDCG@10 for ranking jobs and ranking resumes, respectively.
Abstract:The self-improving ability of large language models (LLMs), enabled by prompting them to analyze and revise their own outputs, has garnered significant interest in recent research. However, this ability has been shown to be absent and difficult to learn for smaller models, thus widening the performance gap between state-of-the-art LLMs and more cost-effective and faster ones. To reduce this gap, we introduce TriPosT, a training algorithm that endows smaller models with such self-improvement ability, and show that our approach can improve a LLaMA-7b's performance on math and reasoning tasks by up to 7.13%. In contrast to prior work, we achieve this by using the smaller model to interact with LLMs to collect feedback and improvements on its own generations. We then replay this experience to train the small model. Our experiments on four math and reasoning datasets show that the interactive experience of learning from and correcting its own mistakes is crucial for small models to improve their performance.
Abstract:Joint entity and relation extraction is a process that identifies entity pairs and their relations using a single model. We focus on the problem of training these models on distantly-labeled data, which is generated by aligning entity mentions in a text corpus with their corresponding entity and relation types in a knowledge base. One key challenge here is the presence of noisy labels, which arises from both entity and relation annotations, and significantly impair the effectiveness of supervised learning applications. However, existing research primarily addresses only one type of noise, thereby limiting the effectiveness of noise reduction. To fill this gap, we introduce a new noise-robust approach, that 1)~incorporates a pre-trained GPT-2 into a sequence tagging scheme for simultaneous entity and relation detection, and 2)~employs a noise-robust learning framework which includes a new loss function that penalizes inconsistency with both significant relation patterns and entity-relation dependencies, as well as a self-adaptive learning step that iteratively selects and trains on high-quality instances. Experiments on two datasets show that our method outperforms the existing state-of-the-art methods in both joint extraction performance and noise reduction effect.
Abstract:Planning for goal-oriented dialogue often requires simulating future dialogue interactions and estimating task progress. Many approaches thus consider training neural networks to perform look-ahead search algorithms such as A* search and Monte Carlo Tree Search (MCTS). However, this training often require abundant annotated data, which creates challenges when faced with noisy annotations or low-resource settings. We introduce GDP-Zero, an approach using Open-Loop MCTS to perform goal-oriented dialogue policy planning without any model training. GDP-Zero prompts a large language model to act as a policy prior, value function, user simulator, and system model during the tree search. We evaluate GDP-Zero on the goal-oriented task PersuasionForGood, and find that its responses are preferred over ChatGPT up to 59.32% of the time, and are rated more persuasive than ChatGPT during interactive evaluations.
Abstract:Large Language Models (LLMs) can generate texts that carry the risk of various misuses, including plagiarism, planting fake reviews on e-commerce platforms, or creating fake social media postings that can sway election results. Detecting whether a text is machine-generated has thus become increasingly important. While machine-learning-based detection strategies exhibit superior performance, they often lack generalizability, limiting their practicality. In this work, we introduce GPT Paternity Test (GPT-Pat), which reliably detects machine-generated text across varied datasets. Given a text under scrutiny, we leverage ChatGPT to generate a corresponding question and provide a re-answer to the question. By comparing the similarity between the original text and the generated re-answered text, it can be determined whether the text is machine-generated. GPT-Pat consists of a Siamese network to compute the similarity between the original text and the generated re-answered text and a binary classifier. Our method achieved an average accuracy of 94.57% on four generalization test sets, surpassing the state-of-the-art RoBERTa-based method by 12.34%. The accuracy drop of our method is only about half of that of the RoBERTa-based method when it is attacked by re-translation and polishing.