arXiv subjects

Fangyu Liu

Publications and source records attributed to Fangyu Liu.

At least 19 recordsLinked to original sources

arXiv2026

Relic Density of Asymmetric Dark Matter with Breit-Wigner Enhancement

The Breit-Wigner Enhancement mechanism provides a novel interpretation for the ``enhancement factor'' derived from data of the PAMELA, ATIC, and PPB-BETS experiments, and we further extends this research to the case of asymmetric Dark Matter. We investigate the impact of resonance mechanism to the asymmetric Dark Matter case and constrain the resonance model parameters to satisfy the ``enhancement factor" and the asymmetric dark matter model parameters to meet the upper bound requirement imposed by the astronomical experiment.

hep-ph↗

arXiv2026

Intention-Aware Semantic Agent Communications for AI Glasses

Smart glasses are emerging as a promising interface between humans and artificial intelligence (AI) agents, enabling first-person perception, contextual awareness, and real-time assistance. However, continuous offloading of visual data from wearable devices to cloud-based vision-language models (VLMs) is fundamentally constrained by limited wireless bandwidth and energy resources. This paper proposes an intention-aware semantic agent communication framework for AI glasses, where data transmission is guided by user intention rather than raw pixel fidelity. In the proposed architecture, AI glasses act as an edge semantic agent while a server-side VLM executes high-level cognition and reasoning. The user intention can be inferred by the server-side VLM through the current transmitted content and the historical prompts. Driven by specific user intentions, the glasses adaptively preserve textual content, document layout, or object semantics before transmission. We evaluate three representative scenarios with different lightweight preprocessing tools on the AI glasses. Simulation results demonstrate that intention-aware preprocessing significantly achieves more than 50% bandwidth reduction depending on the current task while maintaining task performance. Moreover, semantic transmission exhibits graceful degradation under low SNRs. The findings demonstrate that aligning communication resources with user intention is essential for robust and efficient wearable AI agent systems.

eess.SP↗

arXiv2026

Semantic Satellite Communications for Synchronized Audiovisual Reconstruction

Satellite communications face severe bottlenecks in supporting high-fidelity synchronized audiovisual services, as conventional schemes struggle with cross-modal coherence under fluctuating channel conditions, limited bandwidth, and long propagation delays. To address these limitations, this paper proposes an adaptive multimodal semantic transmission system tailored for satellite scenarios, aiming for high-quality synchronized audiovisual reconstruction under bandwidth constraints. Unlike static schemes with fixed modal priorities, our framework features a dual-stream generative architecture that flexibly switches between video-driven audio generation and audio-driven video generation. This allows the system to dynamically decouple semantics, transmitting only the most important modality while employing cross-modal generation to recover the other. To balance reconstruction quality and transmission overhead, a dynamic keyframe update mechanism adaptively maintains the shared knowledge base according to wireless scenarios and user requirements. Furthermore, a large language model based decision module is introduced to enhance system adaptability. By integrating satellite-specific knowledge, this module jointly considers task requirements and channel factors such as weather-induced fading to proactively adjust transmission paths and generation workflows. Simulation results demonstrate that the proposed system significantly reduces bandwidth consumption while achieving high-fidelity audiovisual synchronization, improving transmission efficiency and robustness in challenging satellite scenarios.

eess.IV↗

arXiv2025

Foundation Model-Based Adaptive Semantic Image Transmission for Dynamic Wireless Environments

Foundation model-based semantic transmission has recently shown great potential in wireless image communication. However, existing methods exhibit two major limitations: (i) they overlook the varying importance of semantic components for specific downstream tasks, and (ii) they insufficiently exploit wireless domain knowledge, resulting in limited robustness under dynamic channel conditions. To overcome these challenges, this paper proposes a foundation model-based adaptive semantic image transmission system for dynamic wireless environments, such as autonomous driving. The proposed system decomposes each image into a semantic segmentation map and a compressed representation, enabling task-aware prioritization of critical objects and fine-grained textures. A task-adaptive precoding mechanism then allocates radio resources according to the semantic importance of extracted features. To ensure accurate channel information for precoding, a channel estimation knowledge map (CEKM) is constructed using a conditional diffusion model that integrates user position, velocity, and sparse channel samples to train scenario-specific lightweight estimators. At the receiver, a conditional diffusion model reconstructs high-quality images from the received semantic features, ensuring robustness against channel impairments and partial data loss. Simulation results on the BDD100K dataset with multi-scenario channels generated by QuaDRiGa demonstrate that the proposed method outperforms existing approaches in terms of perceptual quality (SSIM, LPIPS, FID), task-specific accuracy (IoU), and transmission efficiency. These results highlight the effectiveness of integrating task-aware semantic decomposition, scenario-adaptive channel estimation, and diffusion-based reconstruction for robust semantic transmission in dynamic wireless environments.

eess.IV↗

arXiv2025

Constraints on Asymmetric Dark Matter Self Annihilation Cross Sections

We investigate the evolution of the relic density for asymmetric dark matter incorporating self annihilation processes in both the shear-dominated universe and Gauss-Bonnet braneworld. Under the same conditions where the ratio of final asymmetry to the initial asymmetry $ R \equiv \Delta_{-\infty}/\Delta_{\text{in}} $ is identical, the shear-dominated universe, due to its enhanced Hubble expansion rate, leads to an earlier freeze-out point of wash-out asymmetry process and allows a higher upper limit on the self annihilation cross section. Conversely, the Gauss-Bonnet braneworld, with a weakened Hubble expansion rate, delays the freeze-out point and permits a lower upper limit on the self annihilation cross section. We further constrain the wino mass $M_2$ for sneutrino and higgsino asymmetric dark matter in both scenarios, showing that, compared to the standard model, the lower limit of $M_2$ is smaller in the shear-dominated universe but higher in the Gauss-Bonnet braneworld.

hep-ph↗

arXiv2025

Non-standard cosmological scenarios, Sommerfeld enhancement and asymmetric dark matter

We discuss the relic density of asymmetric dark matter with long-range interactions in non-standard cosmological scenarios where the extra cosmic energy is introduced. The Hubble expansion rate is modified in non-standard cosmological models, which affects the relic density of dark matter. If the mass of the dark exchanged gauge boson is less than the mass of dark matter particles multiplied by the dark fine-structure constant, the wave function of the incoming particles of dark matter annihilation would be distorted away from the free plane-wave approximation, yielding the significant enhancements to annihilation cross sections, known as Sommerfeld enhancement. Sommerfeld enhancement results the asymmetric dark matter relic density under abundant, while the enhanced cosmic expansion rate increases the relic density of asymmetric dark matter by letting the decoupling from the thermal equilibrium occurs earlier. However, the mixed effect of modified cosmic expansion and the Sommerfeld enhancement on asymmetric dark matter density evolution becomes quite complicated, due to their opposing affects on relic density and contrary strength developments as the temperature fell. We investigated the mixed effects on asymmetric dark matter freeze-out process and current relic abundance, which is significantly different from the standard situation. Further, we calculate the constraint relations of the asymmetric dark matter pertubative annihilation cross section and dark coupling constant with the mass of asymmetric dark matter, the asymmetry factor with the dark coupling constant, when the relic density of asymmetric dark matter falled in the observed region. The upper bounds on the mass of asymmetric dark matter for s-wave and p-wave annihilations are also derived.

hep-ph↗

arXiv2024

The Hyperfitting Phenomenon: Sharpening and Stabilizing LLMs for Open-Ended Text Generation

This paper introduces the counter-intuitive generalization results of overfitting pre-trained large language models (LLMs) on very small datasets. In the setting of open-ended text generation, it is well-documented that LLMs tend to generate repetitive and dull sequences, a phenomenon that is especially apparent when generating using greedy decoding. This issue persists even with state-of-the-art LLMs containing billions of parameters, trained via next-token prediction on large datasets. We find that by further fine-tuning these models to achieve a near-zero training loss on a small set of samples -- a process we refer to as hyperfitting -- the long-sequence generative capabilities are greatly enhanced. Greedy decoding with these Hyperfitted models even outperform Top-P sampling over long-sequences, both in terms of diversity and human preferences. This phenomenon extends to LLMs of various sizes, different domains, and even autoregressive image generation. We further find this phenomena to be distinctly different from that of Grokking and double descent. Surprisingly, our experiments indicate that hyperfitted models rarely fall into repeating sequences they were trained on, and even explicitly blocking these sequences results in high-quality output. All hyperfitted models produce extremely low-entropy predictions, often allocating nearly all probability to a single token.

cs.CL↗

arXiv2024

Sparkles: Unlocking Chats Across Multiple Images for Multimodal Instruction-Following Models

Large language models exhibit enhanced zero-shot performance on various tasks when fine-tuned with instruction-following data. Multimodal instruction-following models extend these capabilities by integrating both text and images. However, existing models such as MiniGPT-4 and LLaVA face challenges in maintaining dialogue coherence in scenarios involving multiple images. A primary reason is the lack of a specialized dataset for this critical application. To bridge these gaps, we introduce SparklesDialogue, the first machine-generated dialogue dataset tailored for word-level interleaved multi-image and text interactions. Furthermore, we construct SparklesEval, a GPT-assisted benchmark for quantitatively assessing a model's conversational competence across multiple images and dialogue turns. We then present SparklesChat, a multimodal instruction-following model for open-ended dialogues across multiple images. Our experiments validate the effectiveness of training SparklesChat with SparklesDialogue based on MiniGPT-4 and LLaVA-v1.5, which enhances comprehension across multiple images and dialogue turns, and does not compromise single-image understanding capabilities. Qualitative evaluations further demonstrate SparklesChat's generality in handling real-world applications. All resources related to this study are publicly available at https://github.com/HYPJUDY/Sparkles.

cs.CV↗

arXiv2024

Constraints on non-standard cosmological models from Planck data

We review the relic density of dark matter in the non-standard cosmological scenarios which includes kination models, brane world cosmology and shear dominated universe. Then we use the Planck data to find constraints on the parameter spaces as dark matter cross sections and the five dimentional Planck mass for brane cosmology, enhancement factor for kination model and the inverse-scaled shear temperature for shear dominated universe.

hep-ph↗

arXiv2024

Best Practices and Lessons Learned on Synthetic Data

The success of AI models relies on the availability of large, diverse, and high-quality datasets, which can be challenging to obtain due to data scarcity, privacy concerns, and high costs. Synthetic data has emerged as a promising solution by generating artificial data that mimics real-world patterns. This paper provides an overview of synthetic data research, discussing its applications, challenges, and future directions. We present empirical evidence from prior art to demonstrate its effectiveness and highlight the importance of ensuring its factuality, fidelity, and unbiasedness. We emphasize the need for responsible use of synthetic data to build more powerful, inclusive, and trustworthy language models.

cs.CL↗

arXiv2024

PaliGemma: A versatile 3B VLM for transfer

PaliGemma is an open Vision-Language Model (VLM) that is based on the SigLIP-So400m vision encoder and the Gemma-2B language model. It is trained to be a versatile and broadly knowledgeable base model that is effective to transfer. It achieves strong performance on a wide variety of open-world tasks. We evaluate PaliGemma on almost 40 diverse tasks including standard VLM benchmarks, but also more specialized tasks such as remote-sensing and segmentation.

cs.CV↗

arXiv2024

ReMI: A Dataset for Reasoning with Multiple Images

With the continuous advancement of large language models (LLMs), it is essential to create new benchmarks to effectively evaluate their expanding capabilities and identify areas for improvement. This work focuses on multi-image reasoning, an emerging capability in state-of-the-art LLMs. We introduce ReMI, a dataset designed to assess LLMs' ability to Reason with Multiple Images. This dataset encompasses a diverse range of tasks, spanning various reasoning domains such as math, physics, logic, code, table/chart understanding, and spatial and temporal reasoning. It also covers a broad spectrum of characteristics found in multi-image reasoning scenarios. We have benchmarked several cutting-edge LLMs using ReMI and found a substantial gap between their performance and human-level proficiency. This highlights the challenges in multi-image reasoning and the need for further research. Our analysis also reveals the strengths and weaknesses of different models, shedding light on the types of reasoning that are currently attainable and areas where future models require improvement. To foster further research in this area, we are releasing ReMI publicly: https://huggingface.co/datasets/mehrankazemi/ReMI.

cs.CV↗

arXiv2024

Faithful Chart Summarization with ChaTS-Pi

Chart-to-summary generation can help explore data, communicate insights, and help the visually impaired people. Multi-modal generative models have been used to produce fluent summaries, but they can suffer from factual and perceptual errors. In this work we present CHATS-CRITIC, a reference-free chart summarization metric for scoring faithfulness. CHATS-CRITIC is composed of an image-to-text model to recover the table from a chart, and a tabular entailment model applied to score the summary sentence by sentence. We find that CHATS-CRITIC evaluates the summary quality according to human ratings better than reference-based metrics, either learned or n-gram based, and can be further used to fix candidate summaries by removing not supported sentences. We then introduce CHATS-PI, a chart-to-summary pipeline that leverages CHATS-CRITIC during inference to fix and rank sampled candidates from any chart-summarization model. We evaluate CHATS-PI and CHATS-CRITIC using human raters, establishing state-of-the-art results on two popular chart-to-summary datasets.

cs.CL↗

arXiv2024

LUQ: Long-text Uncertainty Quantification for LLMs

Large Language Models (LLMs) have demonstrated remarkable capability in a variety of NLP tasks. However, LLMs are also prone to generate nonfactual content. Uncertainty Quantification (UQ) is pivotal in enhancing our understanding of a model's confidence on its generation, thereby aiding in the mitigation of nonfactual outputs. Existing research on UQ predominantly targets short text generation, typically yielding brief, word-limited responses. However, real-world applications frequently necessitate much longer responses. Our study first highlights the limitations of current UQ methods in handling long text generation. We then introduce \textsc{Luq} and its two variations, a series of novel sampling-based UQ approaches specifically designed for long text. Our findings reveal that \textsc{Luq} outperforms existing baseline methods in correlating with the model's factuality scores (negative coefficient of -0.85 observed for Gemini Pro). To further improve the factuality of LLM responses, we propose \textsc{Luq-Ensemble}, a method that ensembles responses from multiple models and selects the response with the lowest uncertainty. The ensembling method greatly improves the response factuality upon the best standalone LLM.

cs.CL↗

arXiv2024

Chart-based Reasoning: Transferring Capabilities from LLMs to VLMs

Vision-language models (VLMs) are achieving increasingly strong performance on multimodal tasks. However, reasoning capabilities remain limited particularly for smaller VLMs, while those of large-language models (LLMs) have seen numerous improvements. We propose a technique to transfer capabilities from LLMs to VLMs. On the recently introduced ChartQA, our method obtains state-of-the-art performance when applied on the PaLI3-5B VLM by \citet{chen2023pali3}, while also enabling much better performance on PlotQA and FigureQA. We first improve the chart representation by continuing the pre-training stage using an improved version of the chart-to-table translation task by \citet{liu2023deplot}. We then propose constructing a 20x larger dataset than the original training set. To improve general reasoning capabilities and improve numerical operations, we synthesize reasoning traces using the table representation of charts. Lastly, our model is fine-tuned using the multitask loss introduced by \citet{hsieh2023distilling}. Our variant ChartPaLI-5B outperforms even 10x larger models such as PaLIX-55B without using an upstream OCR system, while keeping inference time constant compared to the PaLI3-5B baseline. When rationales are further refined with a simple program-of-thought prompt \cite{chen2023program}, our model outperforms the recently introduced Gemini Ultra and GPT-4V.

cs.CL↗

arXiv2024

Relic abundance of dark matter with coannihilation in non-standard cosmological scenarios

We investigate the relic abundance of dark matter from coannihilation in non-standard cosmological scenarios. We explore the effect of coannihilation on the relic density of dark matter and freeze out temperature in quintessence model with kination phase and brane world cosmological scenarios. Since the Hubble expansion rate is enhanced in quintessence and brane world cosmological models, it causes the larger relic density compared to that in the standard one. On the other hand, the relic density of dark matter is decreased due to the coannihilation in the standard cosmological scenario. After including coannihilation in quintessence or brane world cosmological scenarios, we find the decrease of the relic density of dark matter is slightly slower than that in the standard cosmological scenario.

hep-ph↗

arXiv2024

Constraints on Asymmetric Dark Matter Self Annihilation Cross Sections in Non-standard Cosmological Scenarios

We investigate the relic abundance of asymmetric dark matter in the non-standard cosmological scenarios when the annihilation cross section includes self annihilations. Here we discuss the kination model and brane world cosmology. When the self annihilation is permitted for asymmetric dark matter, there is possibility of washing out the pre-existed asymmetry. We find the constraints on the cross section to avoid the complete washing out of the asymmetry in the non-standard cosmological scenarios. The enhanced cosmic expansion rate causes the freeze out point of wash-out to be earlier. The larger self annihilation cross sections are allowed to exist in kination model and brane world cosmology. Then, in the case of left-handed sneutrino asymmetric dark matter, we find the value of the lower bound on winos mass is smaller than that in the standard cosmological scenario.

hep-ph↗

arXiv2024

On-the-fly Denoising for Data Augmentation in Natural Language Understanding

Data Augmentation (DA) is frequently used to provide additional training data without extra human annotation automatically. However, data augmentation may introduce noisy data that impairs training. To guarantee the quality of augmented data, existing methods either assume no noise exists in the augmented data and adopt consistency training or use simple heuristics such as training loss and diversity constraints to filter out "noisy" data. However, those filtered examples may still contain useful information, and dropping them completely causes a loss of supervision signals. In this paper, based on the assumption that the original dataset is cleaner than the augmented data, we propose an on-the-fly denoising technique for data augmentation that learns from soft augmented labels provided by an organic teacher model trained on the cleaner original data. To further prevent overfitting on noisy labels, a simple self-regularization module is applied to force the model prediction to be consistent across two distinct dropouts. Our method can be applied to general augmentation techniques and consistently improve the performance on both text classification and question-answering tasks.

cs.CL↗