Machine unlearning is an emerging technology that has come to attract widespread attention. A number of factors, including regulations and laws, privacy, and usability concerns, have resulted in this need to allow a trained model to forget some of its training data. Existing studies of machine unlearning mainly focus on unlearning requests that forget a cluster of instances or all instances from one class. While these approaches are effective in removing instances, they do not scale to scenarios where partial targets within an instance need to be forgotten. For example, one would like to only unlearn a person from all instances that simultaneously contain the person and other targets. Directly migrating instance-level unlearning to target-level unlearning will reduce the performance of the model after the unlearning process, or fail to erase information completely. To address these concerns, we have proposed a more effective and efficient unlearning scheme that focuses on removing partial targets from the model, which we name "target unlearning". Specifically, we first construct an essential graph data structure to describe the relationships between all important parameters that are selected based on the model explanation method. After that, we simultaneously filter parameters that are also important for the remaining targets and use the pruning-based unlearning method, which is a simple but effective solution to remove information about the target that needs to be forgotten. Experiments with different training models on various datasets demonstrate the effectiveness of the proposed approach.

In this article, we present a novel information access approach inspired by the information foraging theory (IFT) and elephant herding optimization (EHO). First, we propose a model for information access on social media based on the IFT. We then elaborate an adaptation of the original EHO algorithm to apply it to the information access problem. The combination of the IFT and EHO constitutes a good opportunity to find relevant information on social media. However, when dealing with voluminous data, the performance undergoes a sharp drop. To overcome this issue, we developed an enhanced version of EHO for large scale information access. We introduce new operators to the algorithm, including territories delimitation and clan migration using clustering. To validate our work, we created a dataset of more than 1.4 million tweets, on which we carried out extensive experiments. The outcomes reveal the ability of our approach to find relevant information in an effective and efficient way. They also highlight the advantages of the improved version of EHO over the original algorithm regarding different aspects. Furthermore, we undertook a comparative study with two other metaheuristic-based information foraging approaches, namely ant colony system and particle swarm optimization. Overall, the results are very promising.

Large Language Models (LLMs) are rapidly evolving and impacting various fields, necessitating the development of effective methods to evaluate and compare their performance. Most current approaches for performance evaluation are either based on fixed, domain-specific questions that lack the flexibility required in many real-world applications where tasks are not always from a single domain, or rely on human input, making them unscalable. We propose an automated benchmarking framework based on debates between LLMs, judged by another LLM. This method assesses not only domain knowledge, but also skills such as problem definition and inconsistency recognition. We evaluate the performance of various state-of-the-art LLMs using the debate framework and achieve rankings that align closely with popular rankings based on human input, eliminating the need for costly human crowdsourcing.

Alignment of large language models is a critical process designed to ensure that the model's responses to user prompts accurately reflect human intentions and adhere to societal values. The standard Reinforcement Learning from Human Feedback (RLHF) framework primarily focuses on optimizing the performance of large language models using pre-collected prompts. However, collecting prompts that provide comprehensive coverage is both tedious and challenging, and often fails to include scenarios that LLMs need to improve on the most. In this paper, we investigate alignment through the lens of two-agent games, involving iterative interactions between an adversarial and a defensive agent. The adversarial agent's task at each step is to generate prompts that expose the weakness of the defensive agent. In return, the defensive agent seeks to improve its responses to these newly identified prompts it "struggled" with, based on feedback from the reward model. We theoretically demonstrate that this iterative reinforcement learning optimization converges to a Nash Equilibrium for the game induced by the agents. Experimental results in safety scenarios demonstrate that learning in such a competitive environment not only fully trains agents but also leads to policies with enhanced generalization capabilities for both adversarial and defensive agents. Our code is released at https://github.com/ruizheng20/gpo.

Transparency is an important aspect of human-robot interaction (HRI), as it can improve system trust and usability leading to improved communication and performance. However, most transparency models focus only on the amount of information given to users. In this paper, we propose a bidirectional transparency model, termed a transparency-based action (TBA) model, which in addition to providing transparency information (robot-to-human), allows the robot to take actions based on transparency information received from the human (robot-of-human and human-to-robot). We implemented a three-level (High, Medium and Low) TBA model on a robotic system trainer in two pilot studies (with students as participants) to examine its impact on acceptance and HRI. Based on the pilot studies results, the Medium TBA level was not included in the main experiment, which was conducted with older adults (aged 75-85). In that experiment, two TBA levels were compared: Low (basic information including only robot-to-human transparency) and High (including additional information relating to predicted outcomes with robot-of-human and human-to-robot transparency). The results revealed a significant difference between the two TBA levels of the model in terms of perceived usefulness, ease of use, and attitude. The High TBA level resulted in improved user acceptance and was preferred by the users.

In this past year, large language models (LLMs) have had remarkable success in domains outside the traditional natural language processing, and people are starting to explore the usage of LLMs in more general and close to application domains like code generation, travel planning, and robot controls. Connecting these LLMs with great capacity and external tools, people are building the so-called LLM agents, which are supposed to help people do all kinds of work in everyday life. In all these domains, the prompt to the LLMs has been shown to make a big difference in what the LLM would generate and thus affect the performance of the LLM agents. Therefore, automatic prompt engineering has become an important question for many researchers and users of LLMs. In this paper, we propose a novel method, \textsc{RePrompt}, which does "gradient descent" to optimize the step-by-step instructions in the prompt of the LLM agents based on the chat history obtained from interactions with LLM agents. By optimizing the prompt, the LLM will learn how to plan in specific domains. We have used experiments in PDDL generation and travel planning to show that our method could generally improve the performance for different reasoning tasks when using the updated prompt as the initial prompt.

Tandem mass spectrometry has played a pivotal role in advancing proteomics, enabling the high-throughput analysis of protein composition in biological tissues. Many deep learning methods have been developed for \emph{de novo} peptide sequencing task, i.e., predicting the peptide sequence for the observed mass spectrum. However, two key challenges seriously hinder the further advancement of this important task. Firstly, since there is no consensus for the evaluation datasets, the empirical results in different research papers are often not comparable, leading to unfair comparison. Secondly, the current methods are usually limited to amino acid-level or peptide-level precision and recall metrics. In this work, we present the first unified benchmark NovoBench for \emph{de novo} peptide sequencing, which comprises diverse mass spectrum data, integrated models, and comprehensive evaluation metrics. Recent impressive methods, including DeepNovo, PointNovo, Casanovo, InstaNovo, AdaNovo and $\pi$-HelixNovo are integrated into our framework. In addition to amino acid-level and peptide-level precision and recall, we evaluate the models' performance in terms of identifying post-tranlational modifications (PTMs), efficiency and robustness to peptide length, noise peaks and missing fragment ratio, which are important influencing factors while seldom be considered. Leveraging this benchmark, we conduct a large-scale study of current methods, report many insightful findings that open up new possibilities for future development. The benchmark will be open-sourced to facilitate future research and application.

Federated learning (FL) has obtained tremendous progress in providing collaborative training solutions for distributed data silos with privacy guarantees. However, few existing works explore a more realistic scenario where the clients hold multiple data modalities. In this paper, we aim to solve a novel challenge in multi-modal federated learning (MFL) -- modality missing -- the clients may lose part of the modalities in their local data sets. To tackle the problems, we propose a novel multi-modal federated learning method, Federated Multi-modal contrastiVe training with Pre-trained completion (FedMVP), which integrates the large-scale pre-trained models to enhance the federated training. In the proposed FedMVP framework, each client deploys a large-scale pre-trained model with frozen parameters for modality completion and representation knowledge transfer, enabling efficient and robust local training. On the server side, we utilize generated data to uniformly measure the representation similarity among the uploaded client models and construct a graph perspective to aggregate them according to their importance in the system. We demonstrate that the model achieves superior performance over two real-world image-text classification datasets and is robust to the performance degradation caused by missing modality.

Recently, Multimodal Large Language Models (MLLMs) have been used as agents to control keyboard and mouse inputs by directly perceiving the Graphical User Interface (GUI) and generating corresponding code. However, current agents primarily exhibit excellent understanding capabilities in static environments and are predominantly applied in relatively simple domains, such as Web or mobile interfaces. We argue that a robust GUI agent should be capable of perceiving temporal information on the GUI, including dynamic Web content and multi-step tasks. Additionally, it should possess a comprehensive understanding of various GUI scenarios, including desktop software and multi-window interactions. To this end, this paper introduces a new dataset, termed GUI-World, which features meticulously crafted Human-MLLM annotations, extensively covering six GUI scenarios and eight types of GUI-oriented questions in three formats. We evaluate the capabilities of current state-of-the-art MLLMs, including ImageLLMs and VideoLLMs, in understanding various types of GUI content, especially dynamic and sequential content. Our findings reveal that ImageLLMs struggle with dynamic GUI content without manually annotated keyframes or operation history. On the other hand, VideoLLMs fall short in all GUI-oriented tasks given the sparse GUI video dataset. Based on GUI-World, we take the initial step of leveraging a fine-tuned VideoLLM as a GUI agent, demonstrating an improved understanding of various GUI tasks. However, due to the limitations in the performance of base LLMs, we conclude that using VideoLLMs as GUI agents remains a significant challenge. We believe our work provides valuable insights for future research in dynamic GUI content understanding. The code and dataset are publicly available at our project homepage: https://gui-world.github.io/.

Learning control policies to perform complex robotics tasks from human preference data presents significant challenges. On the one hand, the complexity of such tasks typically requires learning policies to perform a variety of subtasks, then combining them to achieve the overall goal. At the same time, comprehensive, well-engineered reward functions are typically unavailable in such problems, while limited human preference data often is; making efficient use of such data to guide learning is therefore essential. Methods for learning to perform complex robotics tasks from human preference data must overcome both these challenges simultaneously. In this work, we introduce DIPPER: Direct Preference Optimization to Accelerate Primitive-Enabled Hierarchical Reinforcement Learning, an efficient hierarchical approach that leverages direct preference optimization to learn a higher-level policy and reinforcement learning to learn a lower-level policy. DIPPER enjoys improved computational efficiency due to its use of direct preference optimization instead of standard preference-based approaches such as reinforcement learning from human feedback, while it also mitigates the well-known hierarchical reinforcement learning issues of non-stationarity and infeasible subgoal generation due to our use of primitive-informed regularization inspired by a novel bi-level optimization formulation of the hierarchical reinforcement learning problem. To validate our approach, we perform extensive experimental analysis on a variety of challenging robotics tasks, demonstrating that DIPPER outperforms hierarchical and non-hierarchical baselines, while ameliorating the non-stationarity and infeasible subgoal generation issues of hierarchical reinforcement learning.