Diversified retrieval and online learning are two core research areas in the design of modern information retrieval systems. In this paper, we propose the linear submodular bandits problem, which is an online learning setting for optimizing a general class of feature-rich submodular utility models for diversified retrieval.

Progress in machine learning (ML) has been fueled by scaling neural network models. This scaling has been enabled by ever more heroic feats of engineering, necessary for accommodating ML approaches that require high bandwidth communication between devices working in parallel. In this work, we propose a co-designed modular architecture and training approach for ML models, dubbed DIstributed PAth COmposition (DiPaCo). During training, DiPaCo distributes computation by paths through a set of shared modules. Together with a Local-SGD inspired optimization (DiLoCo) that keeps modules in sync with drastically reduced communication, Our approach facilitates training across poorly connected and heterogeneous workers, with a design that ensures robustness to worker failures and preemptions. At inference time, only a single path needs to be executed for each input, without the need for any model compression. We consider this approach as a first prototype towards a new paradigm of large-scale learning, one that is less synchronous and more modular. Our experiments on the widely used C4 benchmark show that, for the same amount of training steps but less wall-clock time, DiPaCo exceeds the performance of a 1 billion-parameter dense transformer language model by choosing one of 256 possible paths, each with a size of 150 million parameters.

Instruction tuning for large language models (LLMs) can drive them to produce results consistent with human goals in specific downstream tasks. However, the process of continual instruction tuning (CIT) for LLMs may bring about the catastrophic forgetting (CF) problem, where previously learned abilities are degraded. Recent methods try to alleviate the CF problem by modifying models or replaying data, which may only remember the surface-level pattern of instructions and get confused on held-out tasks. In this paper, we propose a novel continual instruction tuning method based on Key-part Information Gain (KPIG). Our method computes the information gain on masked parts to dynamically replay data and refine the training objective, which enables LLMs to capture task-aware information relevant to the correct response and alleviate overfitting to general descriptions in instructions. In addition, we propose two metrics, P-score and V-score, to measure the generalization and instruction-following abilities of LLMs. Experiments demonstrate our method achieves superior performance on both seen and held-out tasks.

Distributed nonconvex optimization underpins key functionalities of numerous distributed systems, ranging from power systems, smart buildings, cooperative robots, vehicle networks to sensor networks. Recently, it has also merged as a promising solution to handle the enormous growth in data and model sizes in deep learning. A fundamental problem in distributed nonconvex optimization is avoiding convergence to saddle points, which significantly degrade optimization accuracy. We discover that the process of quantization, which is necessary for all digital communications, can be exploited to enable saddle-point avoidance. More specifically, we propose a stochastic quantization scheme and prove that it can effectively escape saddle points and ensure convergence to a second-order stationary point in distributed nonconvex optimization. With an easily adjustable quantization granularity, the approach allows a user to control the number of bits sent per iteration and, hence, to aggressively reduce the communication overhead. Numerical experimental results using distributed optimization and learning problems on benchmark datasets confirm the effectiveness of the approach.

Many studies have shown how ergonomically designed furniture improves productivity and well-being. As computers have become a part of students' academic lives, they will grow further in the future. We propose anthropometric-based furniture dimensions suitable for university students to improve computer laboratory ergonomics. We collected data from 380 participants and analyzed 11 anthropometric measurements, correlating them to 11 furniture dimensions. Two types of furniture were studied: a non-adjustable chair with a non-adjustable table and an adjustable chair with a non-adjustable table. The mismatch calculation showed a significant difference between furniture dimensions and anthropometric measurements. The one-way ANOVA test with a significance level of 5% also showed a significant difference between proposed and existing furniture dimensions. The proposed dimensions were found to be more compatible and reduced mismatch percentages for both males and females compared to existing furniture. The proposed dimensions of the furniture set with adjustable seat height showed slightly improved results compared to the non-adjustable furniture set. This suggests that the proposed dimensions can improve comfort levels and reduce the risk of musculoskeletal disorders among students. Further studies on the implementation and long-term effects of these proposed dimensions in real-world computer laboratory settings are recommended.

In this multi-task learning study on simultaneous analysis of emotions and their underlying causes in conversational contexts, deep neural network methods were employed to effectively process and train large labeled datasets. However, these approaches are typically limited to conducting context analyses across the entire corpus because they rely on one of the two methods: word- or sentence-level embedding. The former struggles with polysemy and homonyms, whereas the latter causes information loss when processing long sentences. In this study, we overcome the limitations of previous embeddings by utilizing both word- and sentence-level embeddings. Furthermore, we propose the emotion-causality recognition in conversation (ECRC) model, which is based on a novel graph structure, thereby leveraging the strengths of both embedding methods. This model uniquely integrates the bidirectional long short-term memory (Bi-LSTM) and graph neural network (GCN) models for Korean conversation analysis. Compared with models that rely solely on one embedding method, the proposed model effectively structures abstract concepts, such as language features and relationships, thereby minimizing information loss. To assess model performance, we compared the multi-task learning results of three deep neural network models with varying graph structures. Additionally, we evaluated the proposed model using Korean and English datasets. The experimental results show that the proposed model performs better in emotion and causality multi-task learning (74.62% and 75.30%, respectively) when node and edge characteristics are incorporated into the graph structure. Similar results were recorded for the Korean ECC and Wellness datasets (74.62% and 73.44%, respectively) with 71.35% on the IEMOCAP English dataset.

Interactive grasping from clutter, akin to human dexterity, is one of the longest-standing problems in robot learning. Challenges stem from the intricacies of visual perception, the demand for precise motor skills, and the complex interplay between the two. In this work, we present Teacher-Augmented Policy Gradient (TAPG), a novel two-stage learning framework that synergizes reinforcement learning and policy distillation. After training a teacher policy to master the motor control based on object pose information, TAPG facilitates guided, yet adaptive, learning of a sensorimotor policy, based on object segmentation. We zero-shot transfer from simulation to a real robot by using Segment Anything Model for promptable object segmentation. Our trained policies adeptly grasp a wide variety of objects from cluttered scenarios in simulation and the real world based on human-understandable prompts. Furthermore, we show robust zero-shot transfer to novel objects. Videos of our experiments are available at \url{https://maltemosbach.github.io/grasp_anything}.

This study addresses the challenge of forming effective groups in collaborative problem-solving environments. Recognizing the complexity of human interactions and the necessity for efficient collaboration, we propose a novel approach leveraging graph theory and reinforcement learning. Our methodology involves constructing a graph from a dataset where nodes represent participants, and edges signify the interactions between them. We conceptualize each participant as an agent within a reinforcement learning framework, aiming to learn an optimal graph structure that reflects effective group dynamics. Clustering techniques are employed to delineate clear group structures based on the learned graph. Our approach provides theoretical solutions based on evaluation metrics and graph measurements, offering insights into potential improvements in group effectiveness and reductions in conflict incidences. This research contributes to the fields of collaborative work and educational psychology by presenting a data-driven, analytical approach to group formation. It has practical implications for organizational team building, classroom settings, and any collaborative scenario where group dynamics are crucial. The study opens new avenues for exploring the application of graph theory and reinforcement learning in social and behavioral sciences, highlighting the potential for empirical validation in future work.

The rapid adoption of text-to-image diffusion models in society underscores an urgent need to address their biases. Without interventions, these biases could propagate a skewed worldview and restrict opportunities for minority groups. In this work, we frame fairness as a distributional alignment problem. Our solution consists of two main technical contributions: (1) a distributional alignment loss that steers specific characteristics of the generated images towards a user-defined target distribution, and (2) adjusted direct finetuning of diffusion model's sampling process (adjusted DFT), which leverages an adjusted gradient to directly optimize losses defined on the generated images. Empirically, our method markedly reduces gender, racial, and their intersectional biases for occupational prompts. Gender bias is significantly reduced even when finetuning just five soft tokens. Crucially, our method supports diverse perspectives of fairness beyond absolute equality, which is demonstrated by controlling age to a $75\%$ young and $25\%$ old distribution while simultaneously debiasing gender and race. Finally, our method is scalable: it can debias multiple concepts at once by simply including these prompts in the finetuning data. We share code and various fair diffusion model adaptors at https://sail-sg.github.io/finetune-fair-diffusion/.

In this study, we tackle a growing concern around the safety and ethical use of large language models (LLMs). Despite their potential, these models can be tricked into producing harmful or unethical content through various sophisticated methods, including 'jailbreaking' techniques and targeted manipulation. Our work zeroes in on a specific issue: to what extent LLMs can be led astray by asking them to generate responses that are instruction-centric such as a pseudocode, a program or a software snippet as opposed to vanilla text. To investigate this question, we introduce TechHazardQA, a dataset containing complex queries which should be answered in both text and instruction-centric formats (e.g., pseudocodes), aimed at identifying triggers for unethical responses. We query a series of LLMs -- Llama-2-13b, Llama-2-7b, Mistral-V2 and Mistral 8X7B -- and ask them to generate both text and instruction-centric responses. For evaluation we report the harmfulness score metric as well as judgements from GPT-4 and humans. Overall, we observe that asking LLMs to produce instruction-centric responses enhances the unethical response generation by ~2-38% across the models. As an additional objective, we investigate the impact of model editing using the ROME technique, which further increases the propensity for generating undesirable content. In particular, asking edited LLMs to generate instruction-centric responses further increases the unethical response generation by ~3-16% across the different models.