arXiv subjects

Chen Sun

Publications and source records attributed to Chen Sun.

At least 127 records · Page 7Linked to original sources

arXiv2024

Universal scaling of quantum state transport in one-dimensional topological chain under nonadiabatic dynamics

When a system is driven across a continuous phase transition, the density of topological defects demonstrates a power-law scaling behavior versus the quenching rate, as predicted by Kibble-Zurek mechanism. In this study, we generalized this idea and address the scaling of quantum state transport in a one-dimensional topological system subject to a linear drive through its topological quantum phase transition point. We illustrate the power-law dependencies of the quantum state's transport distance, width, and peak magnitude on the driving velocity. Crucially, the power-law exponents are distinct for the edge state and bulk state. Our results offer a novel perspective on quantum state transfer and enriches the field of Kibble-Zurek behaviors and nonadiabatic quantum dynamics.

quant-ph↗

arXiv2024

Solution to a class of multistate Landau-Zener model beyond integrability conditions

We study a class of multistate Landau-Zener model which cannot be solved by integrability conditions or other standard techniques. By analyzing analytical constraints on its scattering matrix and performing fitting to results from numerical simulations of the Schrödinger equation, we find nearly exact analytical expressions of all its transition probabilities for specific parameter choices. We also determine the transition probabilities up to leading orders of series expansions in terms of the inverse sweep rate (namely, in the diabatic limit) for general parameter choices. We further show that this model can describe a Su-Schrieffer-Heeger chain with couplings changing linearly in time. Our work presents a new route, i.e., analytical constraint plus fitting, to analyze those multistate Landau-Zener models which are beyond the applicability of conventional solving methods.

quant-ph↗

arXiv2024

On the Combination of AI and Wireless Technologies: 3GPP Standardization Progress

Combing Artificial Intelligence (AI) and wireless communication technologies has become one of the major technologies trends towards 2030. This includes using AI to improve the efficiency of the wireless transmission and supporting AI deployment with wireless networks. In this article, the latest progress of the Third Generation Partnership Project (3GPP) standards development is introduced. Concentrating on AI model distributed transfer and AI for Beam Management (BM) with wireless network, we introduce the latest studies and explain how the existing standards should be modified to incorporate the results from academia.

cs.NI↗

arXiv2024

Self-Correcting Self-Consuming Loops for Generative Model Training

As synthetic data becomes higher quality and proliferates on the internet, machine learning models are increasingly trained on a mix of human- and machine-generated data. Despite the successful stories of using synthetic data for representation learning, using synthetic data for generative model training creates "self-consuming loops" which may lead to training instability or even collapse, unless certain conditions are met. Our paper aims to stabilize self-consuming generative model training. Our theoretical results demonstrate that by introducing an idealized correction function, which maps a data point to be more likely under the true data distribution, self-consuming loops can be made exponentially more stable. We then propose self-correction functions, which rely on expert knowledge (e.g. the laws of physics programmed in a simulator), and aim to approximate the idealized corrector automatically and at scale. We empirically validate the effectiveness of self-correcting self-consuming loops on the challenging human motion synthesis task, and observe that it successfully avoids model collapse, even when the ratio of synthetic data to real data is as high as 100%.

cs.LG↗

arXiv2024

Multi-Beam Integrated Sensing and Communication: State-of-the-Art, Challenges and Opportunities

Integrated sensing and communication (ISAC) has been envisioned as a critical enabling technology for the next-generation wireless communication, which can realize location/motion detection of surroundings with communication devices. This additional sensing capability leads to a substantial network quality gain and expansion of the service scenarios. As the system evolves to millimeter wave (mmWave) and above, ISAC can realize simultaneous communications and sensing of the ultra-high throughput level and radar resolution with compact design, which relies on directional beamforming against the path loss. With the multi-beam technology, the dual functions of ISAC can be seamlessly incorporated at the beamspace level by unleashing the potential of joint beamforming. To this end, this article investigates the key technologies for multi-beam ISAC system. We begin with an overview of the current state-of-the-art solutions in multi-beam ISAC. Subsequently, a detailed analysis of the advantages associated with the multi-beam ISAC is provided. Additionally, the key technologies for transmitter, channel and receiver of the multi-beam ISAC are introduced. Finally, we explore the challenges and opportunities presented by multi-beam ISAC, offering valuable insights into this emerging field.

eess.SP↗

arXiv2024

Sterile neutrino dark matter within the $ν$SMEFT

Sterile neutrinos with masses at the $\mathrm{keV}$ scale and mixing to the active neutrinos offer an elegant explanation of the observed dark matter (DM) density. However, the very same mixing inevitably leads to radiative photon emission and the non-observation of such peaked $X$-ray lines rules out this minimal sterile neutrino DM hypothesis. We show that in the context of the Standard Model effective field theory with sterile neutrinos ($ν$SMEFT), higher dimensional operators can produce sterile neutrino DM in a broad range of parameter space. In particular, $ν$SMEFT interactions can open the large mixing parameter space due to their destructive interference, through operator mixing or matching, in the $X$-ray emission. We also find that, even in the zero mixing limit, the DM density can always be explained by $ν$SMEFT operators. The testability of the studied $ν$SMEFT operators in searches for electric dipole moments, neutrinoless double beta decay, and pion decay measurements is discussed.

hep-ph↗

arXiv2024

AntGPT: Can Large Language Models Help Long-term Action Anticipation from Videos?

Can we better anticipate an actor's future actions (e.g. mix eggs) by knowing what commonly happens after his/her current action (e.g. crack eggs)? What if we also know the longer-term goal of the actor (e.g. making egg fried rice)? The long-term action anticipation (LTA) task aims to predict an actor's future behavior from video observations in the form of verb and noun sequences, and it is crucial for human-machine interaction. We propose to formulate the LTA task from two perspectives: a bottom-up approach that predicts the next actions autoregressively by modeling temporal dynamics; and a top-down approach that infers the goal of the actor and plans the needed procedure to accomplish the goal. We hypothesize that large language models (LLMs), which have been pretrained on procedure text data (e.g. recipes, how-tos), have the potential to help LTA from both perspectives. It can help provide the prior knowledge on the possible next actions, and infer the goal given the observed part of a procedure, respectively. To leverage the LLMs, we propose a two-stage framework, AntGPT. It first recognizes the actions already performed in the observed videos and then asks an LLM to predict the future actions via conditioned generation, or to infer the goal and plan the whole procedure by chain-of-thought prompting. Empirical results on the Ego4D LTA v1 and v2 benchmarks, EPIC-Kitchens-55, as well as EGTEA GAZE+ demonstrate the effectiveness of our proposed approach. AntGPT achieves state-of-the-art performance on all above benchmarks, and can successfully infer the goal and thus perform goal-conditioned "counterfactual" prediction via qualitative analysis. Code and model will be released at https://brown-palm.github.io/AntGPT

cs.CV↗

arXiv2024

Onsager Relations between Spin Currents and Charge Currents

We consider the macroscopic dynamics of systems with charge and spin currents, using the methods of Onsager's irreversible thermodynamics. Applied to systems with spin-orbit interaction (SOI), we derive Onsager relations showing that, if electrical disequilibrium leads to spin currents, then magnetic disequilibrium leads to charge currents. We consider three examples of such SOI. Two of these predicted charge currents have not previously appeared. By measuring these charge currents one can infer the corresponding spin currents.

cond-mat.mes-hall↗

arXiv2023

Does Visual Pretraining Help End-to-End Reasoning?

We aim to investigate whether end-to-end learning of visual reasoning can be achieved with general-purpose neural networks, with the help of visual pretraining. A positive result would refute the common belief that explicit visual abstraction (e.g. object detection) is essential for compositional generalization on visual reasoning, and confirm the feasibility of a neural network "generalist" to solve visual recognition and reasoning tasks. We propose a simple and general self-supervised framework which "compresses" each video frame into a small set of tokens with a transformer network, and reconstructs the remaining frames based on the compressed temporal context. To minimize the reconstruction loss, the network must learn a compact representation for each image, as well as capture temporal dynamics and object permanence from temporal context. We perform evaluation on two visual reasoning benchmarks, CATER and ACRE. We observe that pretraining is essential to achieve compositional generalization for end-to-end visual reasoning. Our proposed framework outperforms traditional supervised pretraining, including image classification and explicit object detection, by large margins.

cs.CV↗

arXiv2023

Pixel Aligned Language Models

Large language models have achieved great success in recent years, so as their variants in vision. Existing vision-language models can describe images in natural languages, answer visual-related questions, or perform complex reasoning about the image. However, it is yet unclear how localization tasks, such as word grounding or referring localization, can be performed using large language models. In this work, we aim to develop a vision-language model that can take locations, for example, a set of points or boxes, as either inputs or outputs. When taking locations as inputs, the model performs location-conditioned captioning, which generates captions for the indicated object or region. When generating locations as outputs, our model regresses pixel coordinates for each output word generated by the language model, and thus performs dense word grounding. Our model is pre-trained on the Localized Narrative dataset, which contains pixel-word-aligned captioning from human attention. We show our model can be applied to various location-aware vision-language tasks, including referring localization, location-conditioned captioning, and dense object captioning, archiving state-of-the-art performance on RefCOCO and Visual Genome. Project page: https://jerryxu.net/PixelLLM .

cs.CV↗

arXiv2023

Towards Generic Anomaly Detection and Understanding: Large-scale Visual-linguistic Model (GPT-4V) Takes the Lead

Anomaly detection is a crucial task across different domains and data types. However, existing anomaly detection models are often designed for specific domains and modalities. This study explores the use of GPT-4V(ision), a powerful visual-linguistic model, to address anomaly detection tasks in a generic manner. We investigate the application of GPT-4V in multi-modality, multi-domain anomaly detection tasks, including image, video, point cloud, and time series data, across multiple application areas, such as industrial, medical, logical, video, 3D anomaly detection, and localization tasks. To enhance GPT-4V's performance, we incorporate different kinds of additional cues such as class information, human expertise, and reference images as prompts.Based on our experiments, GPT-4V proves to be highly effective in detecting and explaining global and fine-grained semantic patterns in zero/one-shot anomaly detection. This enables accurate differentiation between normal and abnormal instances. Although we conducted extensive evaluations in this study, there is still room for future evaluation to further exploit GPT-4V's generic anomaly detection capacity from different aspects. These include exploring quantitative metrics, expanding evaluation benchmarks, incorporating multi-round interactions, and incorporating human feedback loops. Nevertheless, GPT-4V exhibits promising performance in generic anomaly detection and understanding, thus opening up a new avenue for anomaly detection.

cs.CV↗

arXiv2023

Towards A Unified Neural Architecture for Visual Recognition and Reasoning

Recognition and reasoning are two pillars of visual understanding. However, these tasks have an imbalance in focus; whereas recent advances in neural networks have shown strong empirical performance in visual recognition, there has been comparably much less success in solving visual reasoning. Intuitively, unifying these two tasks under a singular framework is desirable, as they are mutually dependent and beneficial. Motivated by the recent success of multi-task transformers for visual recognition and language understanding, we propose a unified neural architecture for visual recognition and reasoning with a generic interface (e.g., tokens) for both. Our framework enables the principled investigation of how different visual recognition tasks, datasets, and inductive biases can help enable spatiotemporal reasoning capabilities. Noticeably, we find that object detection, which requires spatial localization of individual objects, is the most beneficial recognition task for reasoning. We further demonstrate via probing that implicit object-centric representations emerge automatically inside our framework. Intriguingly, we discover that certain architectural choices such as the backbone model of the visual encoder have a significant impact on visual reasoning, but little on object detection. Given the results of our experiments, we believe that visual reasoning should be considered as a first-class citizen alongside visual recognition, as they are strongly correlated but benefit from potentially different design choices.

cs.CV↗

arXiv2023

Analyzing Modular Approaches for Visual Question Decomposition

Modular neural networks without additional training have recently been shown to surpass end-to-end neural networks on challenging vision-language tasks. The latest such methods simultaneously introduce LLM-based code generation to build programs and a number of skill-specific, task-oriented modules to execute them. In this paper, we focus on ViperGPT and ask where its additional performance comes from and how much is due to the (state-of-art, end-to-end) BLIP-2 model it subsumes vs. additional symbolic components. To do so, we conduct a controlled study (comparing end-to-end, modular, and prompting-based methods across several VQA benchmarks). We find that ViperGPT's reported gains over BLIP-2 can be attributed to its selection of task-specific modules, and when we run ViperGPT using a more task-agnostic selection of modules, these gains go away. Additionally, ViperGPT retains much of its performance if we make prominent alterations to its selection of modules: e.g. removing or retaining only BLIP-2. Finally, we compare ViperGPT against a prompting-based decomposition strategy and find that, on some benchmarks, modular approaches significantly benefit by representing subtasks with natural language, instead of code.

cs.CV↗

arXiv2023

Emergence of Abstract State Representations in Embodied Sequence Modeling

Decision making via sequence modeling aims to mimic the success of language models, where actions taken by an embodied agent are modeled as tokens to predict. Despite their promising performance, it remains unclear if embodied sequence modeling leads to the emergence of internal representations that represent the environmental state information. A model that lacks abstract state representations would be liable to make decisions based on surface statistics which fail to generalize. We take the BabyAI environment, a grid world in which language-conditioned navigation tasks are performed, and build a sequence modeling Transformer, which takes a language instruction, a sequence of actions, and environmental observations as its inputs. In order to investigate the emergence of abstract state representations, we design a "blindfolded" navigation task, where only the initial environmental layout, the language instruction, and the action sequence to complete the task are available for training. Our probing results show that intermediate environmental layouts can be reasonably reconstructed from the internal activations of a trained model, and that language instructions play a role in the reconstruction accuracy. Our results suggest that many key features of state representations can emerge via embodied sequence modeling, supporting an optimistic outlook for applications of sequence modeling objectives to more complex embodied decision-making domains.

cs.LG↗

arXiv2023

AVIS: Autonomous Visual Information Seeking with Large Language Model Agent

In this paper, we propose an autonomous information seeking visual question answering framework, AVIS. Our method leverages a Large Language Model (LLM) to dynamically strategize the utilization of external tools and to investigate their outputs, thereby acquiring the indispensable knowledge needed to provide answers to the posed questions. Responding to visual questions that necessitate external knowledge, such as "What event is commemorated by the building depicted in this image?", is a complex task. This task presents a combinatorial search space that demands a sequence of actions, including invoking APIs, analyzing their responses, and making informed decisions. We conduct a user study to collect a variety of instances of human decision-making when faced with this task. This data is then used to design a system comprised of three components: an LLM-powered planner that dynamically determines which tool to use next, an LLM-powered reasoner that analyzes and extracts key information from the tool outputs, and a working memory component that retains the acquired information throughout the process. The collected user behavior serves as a guide for our system in two key ways. First, we create a transition graph by analyzing the sequence of decisions made by users. This graph delineates distinct states and confines the set of actions available at each state. Second, we use examples of user decision-making to provide our LLM-powered planner and reasoner with relevant contextual instances, enhancing their capacity to make informed decisions. We show that AVIS achieves state-of-the-art results on knowledge-intensive visual question answering benchmarks such as Infoseek and OK-VQA.

cs.CV↗

arXiv2023

Object-centric Video Representation for Long-term Action Anticipation

This paper focuses on building object-centric representations for long-term action anticipation in videos. Our key motivation is that objects provide important cues to recognize and predict human-object interactions, especially when the predictions are longer term, as an observed "background" object could be used by the human actor in the future. We observe that existing object-based video recognition frameworks either assume the existence of in-domain supervised object detectors or follow a fully weakly-supervised pipeline to infer object locations from action labels. We propose to build object-centric video representations by leveraging visual-language pretrained models. This is achieved by "object prompts", an approach to extract task-specific object-centric representations from general-purpose pretrained models without finetuning. To recognize and predict human-object interactions, we use a Transformer-based neural architecture which allows the "retrieval" of relevant objects for action anticipation at various time scales. We conduct extensive evaluations on the Ego4D, 50Salads, and EGTEA Gaze+ benchmarks. Both quantitative and qualitative results confirm the effectiveness of our proposed method.

cs.CV↗

arXiv2023

Contrastive Retrospection: honing in on critical steps for rapid learning and generalization in RL

In real life, success is often contingent upon multiple critical steps that are distant in time from each other and from the final reward. These critical steps are challenging to identify with traditional reinforcement learning (RL) methods that rely on the Bellman equation for credit assignment. Here, we present a new RL algorithm that uses offline contrastive learning to hone in on these critical steps. This algorithm, which we call Contrastive Retrospection (ConSpec), can be added to any existing RL algorithm. ConSpec learns a set of prototypes for the critical steps in a task by a novel contrastive loss and delivers an intrinsic reward when the current state matches one of the prototypes. The prototypes in ConSpec provide two key benefits for credit assignment: (i) They enable rapid identification of all the critical steps. (ii) They do so in a readily interpretable manner, enabling out-of-distribution generalization when sensory features are altered. Distinct from other contemporary RL approaches to credit assignment, ConSpec takes advantage of the fact that it is easier to retrospectively identify the small set of steps that success is contingent upon (and ignoring other states) than it is to prospectively predict reward at every taken step. ConSpec greatly improves learning in a diverse set of RL tasks. The code is available at the link: https://github.com/sunchipsster1/ConSpec

cs.LG↗

arXiv2023

Goal-Conditioned Predictive Coding for Offline Reinforcement Learning

Recent work has demonstrated the effectiveness of formulating decision making as supervised learning on offline-collected trajectories. Powerful sequence models, such as GPT or BERT, are often employed to encode the trajectories. However, the benefits of performing sequence modeling on trajectory data remain unclear. In this work, we investigate whether sequence modeling has the ability to condense trajectories into useful representations that enhance policy learning. We adopt a two-stage framework that first leverages sequence models to encode trajectory-level representations, and then learns a goal-conditioned policy employing the encoded representations as its input. This formulation allows us to consider many existing supervised offline RL methods as specific instances of our framework. Within this framework, we introduce Goal-Conditioned Predictive Coding (GCPC), a sequence modeling objective that yields powerful trajectory representations and leads to performant policies. Through extensive empirical evaluations on AntMaze, FrankaKitchen and Locomotion environments, we observe that sequence modeling can have a significant impact on challenging decision making tasks. Furthermore, we demonstrate that GCPC learns a goal-conditioned latent representation encoding the future trajectory, which enables competitive performance on all three benchmarks.

cs.LG↗