Tariq Iqbal

Riashat Islam, Hongyu Zang, Manan Tomar et al. · mila

Several self-supervised representation learning methods have been proposed for reinforcement learning (RL) with rich observations. For real-world applications of RL, recovering underlying latent states is crucial, particularly when sensory inputs contain irrelevant and exogenous information. In this work, we study how information bottlenecks can be used to construct latent states efficiently in the presence of task-irrelevant information. We propose architectures that utilize variational and discrete information bottlenecks, coined as RepDIB, to learn structured factorized representations. Exploiting the expressiveness bought by factorized representations, we introduce a simple, yet effective, bottleneck that can be integrated with any existing self-supervised objective for RL. We demonstrate this across several online and offline RL benchmarks, along with a real robot arm task, where we find that compressed representations with RepDIB can lead to strong performance improvements, as the learned bottlenecks help predict only the relevant state while ignoring irrelevant information.

Md Mubtasim Ahasan, Md Fahim, Tasnim Mohiuddin et al.

Recent advancements in speech-language models have yielded significant improvements in speech tokenization and synthesis. However, effectively mapping the complex, multidimensional attributes of speech into discrete tokens remains challenging. This process demands acoustic, semantic, and contextual information for precise speech representations. Existing speech representations generally fall into two categories: acoustic tokens from audio codecs and semantic tokens from speech self-supervised learning models. Although recent efforts have unified acoustic and semantic tokens for improved performance, they overlook the crucial role of contextual representation in comprehensive speech modeling. Our empirical investigations reveal that the absence of contextual representations results in elevated Word Error Rate (WER) and Word Information Lost (WIL) scores in speech transcriptions. To address these limitations, we propose two novel distillation approaches: (1) a language model (LM)-guided distillation method that incorporates contextual information, and (2) a combined LM and self-supervised speech model (SM)-guided distillation technique that effectively distills multimodal representations (acoustic, semantic, and contextual) into a comprehensive speech tokenizer, termed DM-Codec. The DM-Codec architecture adopts a streamlined encoder-decoder framework with a Residual Vector Quantizer (RVQ) and incorporates the LM and SM during the training process. Experiments show DM-Codec significantly outperforms state-of-the-art speech tokenization models, reducing WER by up to 13.46%, WIL by 9.82%, and improving speech quality by 5.84% and intelligibility by 1.85% on the LibriSpeech benchmark dataset. Code, samples, and checkpoints are available at https://github.com/mubtasimahasan/DM-Codec.

Md Mubtasim Ahasan, Rafat Hasan Khan, Tasnim Mohiuddin et al.

Speech tokenization enables discrete representation and facilitates speech language modeling. However, existing neural codecs capture low-level acoustic features, overlooking the semantic and contextual cues inherent to human speech. While recent efforts introduced semantic representations from self-supervised speech models or incorporated contextual representations from pre-trained language models, challenges remain in aligning and unifying the semantic and contextual representations. We introduce FuseCodec, which unifies acoustic, semantic, and contextual representations through strong cross-modal alignment and globally informed supervision. We propose three complementary techniques: (i) Latent Representation Fusion, integrating semantic and contextual features directly into the encoder latent space for robust and unified representation learning; (ii) Global Semantic-Contextual Supervision, supervising discrete tokens with globally pooled and broadcasted representations to enhance temporal consistency and cross-modal alignment; and (iii) Temporally Aligned Contextual Supervision, strengthening alignment by dynamically matching contextual and speech tokens within a local window for fine-grained token-level supervision. We further introduce FuseCodec-TTS, demonstrating our methodology's applicability to zero-shot speech synthesis. Empirically, FuseCodec achieves state-of-the-art performance in LibriSpeech, surpassing EnCodec, SpeechTokenizer, and DAC in transcription accuracy, perceptual quality, intelligibility, and speaker similarity. Results highlight the effectiveness of contextually and semantically guided tokenization for speech tokenization and downstream tasks. Code and pretrained models are available at https://github.com/mubtasimahasan/FuseCodec.

Shaid Hasan, Breenice Lee, Sujan Sarker et al.

Multi-robot systems hold significant promise for social environments such as homes and hospitals, yet existing multi-robot works treat robots as functionally identical, overlooking how robots individual identity shape user perception and how coordination shapes multi-robot behavior when such individuality is present. To address this, we introduce M2HRI, a multimodal multi-agent framework built on large language models that equips each robot with distinct personality and long-term memory, alongside a coordination mechanism conditioned on these differences. In a controlled user study (n = 105) in a multi-agent human-robot interaction (HRI) scenario, we find that LLM-driven personality traits are significantly distinguishable and enhance interaction quality, long-term memory improves personalization and preference awareness, and centralized coordination significantly reduces overlap while improving overall interaction quality. Together, these results demonstrate that both agent individuality and structured coordination are essential for coherent and socially appropriate multi-agent HRI. Project website and code are available at https://project-m2hri.github.io/.

Shaid Hasan, Breenice Lee, Sujan Sarker et al.

Human-robot interaction is increasingly moving toward multi-robot, socially grounded environments. Existing systems struggle to integrate multimodal perception, embodied expression, and coordinated decision-making in a unified framework. This limits natural and scalable interaction in shared physical spaces. We address this gap by introducing a multimodal framework for human-multi-agent interaction in which each robot operates as an autonomous cognitive agent with integrated multimodal perception and Large Language Model (LLM)-driven planning grounded in embodiment. At the team level, a centralized coordination mechanism regulates turn-taking and agent participation to prevent overlapping speech and conflicting actions. Implemented on two humanoid robots, our framework enables coherent multi-agent interaction through interaction policies that combine speech, gesture, gaze, and locomotion. Representative interaction runs demonstrate coordinated multimodal reasoning across agents and grounded embodied responses. Future work will focus on larger-scale user studies and deeper exploration of socially grounded multi-agent interaction dynamics.

Alexi Gladstone, Ganesh Nanduru, Md Mofijul Islam et al.

Inference-time computation techniques, analogous to human System 2 Thinking, have recently become popular for improving model performances. However, most existing approaches suffer from several limitations: they are modality-specific (e.g., working only in text), problem-specific (e.g., verifiable domains like math and coding), or require additional supervision/training on top of unsupervised pretraining (e.g., verifiers or verifiable rewards). In this paper, we ask the question "Is it possible to generalize these System 2 Thinking approaches, and develop models that learn to think solely from unsupervised learning?" Interestingly, we find the answer is yes, by learning to explicitly verify the compatibility between inputs and candidate-predictions, and then re-framing prediction problems as optimization with respect to this verifier. Specifically, we train Energy-Based Transformers (EBTs) -- a new class of Energy-Based Models (EBMs) -- to assign an energy value to every input and candidate-prediction pair, enabling predictions through gradient descent-based energy minimization until convergence. Across both discrete (text) and continuous (visual) modalities, we find EBTs scale faster than the dominant Transformer++ approach during training, achieving an up to 35% higher scaling rate with respect to data, batch size, parameters, FLOPs, and depth. During inference, EBTs improve performance with System 2 Thinking by 29% more than the Transformer++ on language tasks, and EBTs outperform Diffusion Transformers on image denoising while using fewer forward passes. Further, we find that EBTs achieve better results than existing models on most downstream tasks given the same or worse pretraining performance, suggesting that EBTs generalize better than existing approaches. Consequently, EBTs are a promising new paradigm for scaling both the learning and thinking capabilities of models.

Md Akil Raihan Iftee, Mir Sazzat Hossain, Rakibul Hasan Rajib et al.

Continual Test-Time Adaptation (CTTA) is crucial for deploying models in real-world applications with unseen, evolving target domains. Existing CTTA methods, however, often rely on source data or prototypes, limiting their applicability in privacy-sensitive and resource-constrained settings. Additionally, these methods suffer from long-term forgetting, which degrades performance on previously encountered domains as target domains shift. To address these challenges, we propose SloMo-Fast, a source-free, dual-teacher CTTA framework designed for enhanced adaptability and generalization. It includes two complementary teachers: the Slow-Teacher, which exhibits slow forgetting and retains long-term knowledge of previously encountered domains to ensure robust generalization, and the Fast-Teacher rapidly adapts to new domains while accumulating and integrating knowledge across them. This framework preserves knowledge of past domains and adapts efficiently to new ones. We also introduce Cyclic Test-Time Adaptation (Cyclic-TTA), a novel CTTA benchmark that simulates recurring domain shifts. Our extensive experiments demonstrate that SloMo-Fast consistently outperforms state-of-the-art methods across Cyclic-TTA, as well as ten other CTTA settings, highlighting its ability to both adapt and generalize across evolving and revisited domains.

Alexi Gladstone, Ganesh Nanduru, Md Mofijul Islam et al.

One of the predominant methods for training world models is autoregressive prediction in the output space of the next element of a sequence. In Natural Language Processing (NLP), this takes the form of Large Language Models (LLMs) predicting the next token; in Computer Vision (CV), this takes the form of autoregressive models predicting the next frame/token/pixel. However, this approach differs from human cognition in several respects. First, human predictions about the future actively influence internal cognitive processes. Second, humans naturally evaluate the plausibility of predictions regarding future states. Based on this capability, and third, by assessing when predictions are sufficient, humans allocate a dynamic amount of time to make a prediction. This adaptive process is analogous to System 2 thinking in psychology. All these capabilities are fundamental to the success of humans at high-level reasoning and planning. Therefore, to address the limitations of traditional autoregressive models lacking these human-like capabilities, we introduce Energy-Based World Models (EBWM). EBWM involves training an Energy-Based Model (EBM) to predict the compatibility of a given context and a predicted future state. In doing so, EBWM enables models to achieve all three facets of human cognition described. Moreover, we developed a variant of the traditional autoregressive transformer tailored for Energy-Based models, termed the Energy-Based Transformer (EBT). Our results demonstrate that EBWM scales better with data and GPU Hours than traditional autoregressive transformers in CV, and that EBWM offers promising early scaling in NLP. Consequently, this approach offers an exciting path toward training future models capable of System 2 thinking and intelligently searching across state spaces.

Mohammad Samin Yasar, Tariq Iqbal

Human motion prediction is an essential component for enabling closer human-robot collaboration. The task of accurately predicting human motion is non-trivial. It is compounded by the variability of human motion, both at a skeletal level due to the varying size of humans and at a motion level due to individual movement's idiosyncrasies. These variables make it challenging for learning algorithms to obtain a general representation that is robust to the diverse spatio-temporal patterns of human motion. In this work, we propose a modular sequence learning approach that allows end-to-end training while also having the flexibility of being fine-tuned. Our approach relies on the diversity of training samples to first learn a robust representation, which can then be fine-tuned in a continual learning setup to predict the motion of new subjects. We evaluated the proposed approach by comparing its performance against state-of-the-art baselines. The results suggest that our approach outperforms other methods over all the evaluated temporal horizons, using a small amount of data for fine-tuning. The improved performance of our approach opens up the possibility of using continual learning for personalized and reliable motion prediction.

Md Mofijul Islam, Tariq Iqbal

To fluently collaborate with people, robots need the ability to recognize human activities accurately. Although modern robots are equipped with various sensors, robust human activity recognition (HAR) still remains a challenging task for robots due to difficulties related to multimodal data fusion. To address these challenges, in this work, we introduce a deep neural network-based multimodal HAR algorithm, HAMLET. HAMLET incorporates a hierarchical architecture, where the lower layer encodes spatio-temporal features from unimodal data by adopting a multi-head self-attention mechanism. We develop a novel multimodal attention mechanism for disentangling and fusing the salient unimodal features to compute the multimodal features in the upper layer. Finally, multimodal features are used in a fully connect neural-network to recognize human activities. We evaluated our algorithm by comparing its performance to several state-of-the-art activity recognition algorithms on three human activity datasets. The results suggest that HAMLET outperformed all other evaluated baselines across all datasets and metrics tested, with the highest top-1 accuracy of 95.12% and 97.45% on the UTD-MHAD [1] and the UT-Kinect [2] datasets respectively, and F1-score of 81.52% on the UCSD-MIT [3] dataset. We further visualize the unimodal and multimodal attention maps, which provide us with a tool to interpret the impact of attention mechanisms concerning HAR.

Tariq Iqbal, Samantha Rack, Laurel D. Riek

In order to be effective teammates, robots need to be able to understand high-level human behavior to recognize, anticipate, and adapt to human motion. We have designed a new approach to enable robots to perceive human group motion in real-time, anticipate future actions, and synthesize their own motion accordingly. We explore this within the context of joint action, where humans and robots move together synchronously. In this paper, we present an anticipation method which takes high-level group behavior into account. We validate the method within a human-robot interaction scenario, where an autonomous mobile robot observes a team of human dancers, and then successfully and contingently coordinates its movements to "join the dance". We compared the results of our anticipation method to move the robot with another method which did not rely on high-level group behavior, and found our method performed better both in terms of more closely synchronizing the robot's motion to the team, and also exhibiting more contingent and fluent motion. These findings suggest that the robot performs better when it has an understanding of high-level group behavior than when it does not. This work will help enable others in the robotics community to build more fluent and adaptable robots in the future.