Publications (58)
Re-evaluation of bottleneck effect via a coupled monolayer WS_2/photonic crystal heterostructure
Jiaru Zhou, Wenze Lan, Hao Li +6
Exciton-polariton condensates is an important type of Bose-Einstein condensate whose realization requires efficient relaxation of polaritons to the band-energy minima. However, thi…
EXaMCaP: Subset Selection with Entropy Gain Maximization for Probing Capability Gains of Large Chart Understanding Training Sets
Jiapeng Liu, Liang Li, Bing Li +5
Recent works focus on synthesizing Chart Understanding (ChartU) training sets to inject advanced chart knowledge into Multimodal Large Language Models (MLLMs), where the sufficienc…
Representing Nonterminating Rewriting with
Peng Fu
We specify a second-order type system that is tailored for representing nonterminations. The nonterminating trace of a term in a rewrite system …
Think out Loud: Emotion Deducing Explanation in Dialogues
Jiangnan Li, Zheng Lin, Lanrui Wang +6
Humans convey emotions through daily dialogues, making emotion understanding a crucial step of affective intelligence. To understand emotions in dialogues, machines are asked to re…
Advantageous Parameter Expansion Training Makes Better Large Language Models
Naibin Gu, Yilong Chen, Zhenyu Zhang +7
Although scaling up the number of trainable parameters in both pre-training and fine-tuning can effectively improve the performance of large language models, it also leads to incre…
A dressed polarizability framework for interface-coupled meta-atoms and large-scale metasurfaces
Peng Fu, Jean-Paul Hugonin, Maxime Bertrand +2
The modeling of large collections of interacting meta-atoms remains a central challenge in photonics because it requires the simultaneous treatment of complex scatterer geometries,…
EasyVideoR1: Easier RL for Video Understanding
Chuanyu Qin, Chenxu Yang, Qingyi Si +6
Reinforcement learning from verifiable rewards (RLVR) has demonstrated remarkable effectiveness in improving the reasoning capabilities of large language models. As models evolve i…
Object Attribute Matters in Visual Question Answering
Peize Li, Qingyi Si, Peng Fu +2
Visual question answering is a multimodal task that requires the joint comprehension of visual and textual information. However, integrating visual and textual semantics solely thr…
A tutorial introduction to quantum circuit programming in dependently typed Proto-Quipper
Peng Fu, Kohei Kishida, Neil J. Ross +1
We introduce dependently typed Proto-Quipper, or Proto-Quipper-D for short, an experimental quantum circuit programming language with linear dependent types. We give several exampl…
Neutral Utterances are Also Causes: Enhancing Conversational Causal Emotion Entailment with Social Commonsense Knowledge
Jiangnan Li, Fandong Meng, Zheng Lin +5
Conversational Causal Emotion Entailment aims to detect causal utterances for a non-neutral targeted utterance from a conversation. In this work, we build conversations as graphs t…
Adapt Once, Thrive with Updates: Transferable Parameter-Efficient Fine-Tuning on Evolving Base Models
Naibin Gu, Peng Fu, Xiyu Liu +3
Parameter-efficient fine-tuning (PEFT) has become a common method for fine-tuning large language models, where a base model can serve multiple users through PEFT module switching.…
A Win-win Deal: Towards Sparse and Robust Pre-trained Language Models
Yuanxin Liu, Fandong Meng, Zheng Lin +5
Despite the remarkable success of pre-trained language models (PLMs), they still face two challenges: First, large-scale PLMs are inefficient in terms of memory footprint and compu…
Compressing And Debiasing Vision-Language Pre-Trained Models for Visual Question Answering
Qingyi Si, Yuanxin Liu, Zheng Lin +2
Despite the excellent performance of vision-language pre-trained models (VLPs) on conventional VQA task, they still suffer from two problems: First, VLPs tend to rely on language b…
A Biset-Enriched Categorical Model for Proto-Quipper with Dynamic Lifting
Peng Fu, Kohei Kishida, Neil J. Ross +1
Quipper and Proto-Quipper are a family of quantum programming languages that, by their nature as circuit description languages, involve two runtimes: one at which the program gener…
Irrelevance, Heterogeneous Equality, and Call-by-value Dependent Type Systems
Vilhelm Sjöberg, Chris Casinghino, Ki Yung Ahn +7
We present a full-spectrum dependently typed core language which includes both nontermination and computational irrelevance (a.k.a. erasure), a combination which has not been studi…
On the Lambek embedding and the category of product-preserving presheaves
Peng Fu, Kohei Kishida, Neil J. Ross +1
It is well-known that the category of presheaf functors is complete and cocomplete, and that the Yoneda embedding into the presheaf category preserves products. However, the Yoneda…
Towards Conditional Feature Alignment for Cross-Domain Counting
Zhuonan Liang, Dongnan Liu, Jianan Fan +6
Object counting models often degrade under cross-domain deployment because density composition varies across domains and is itself task-relevant. Standard feature alignment methods…
Superconductivity and Electronic Structures of Nickelate Thin Film Superstructures
Zihao Nie, Yueying Li, Wei Lv +14
Ruddlesden-Popper (RP) nickelates have emerged as a crucial platform for exploring the mechanisms of high-temperature superconductivity. However, the Fermi surface topology require…
Reconstruction of Differentially Private Text Sanitization via Large Language Models
Shuchao Pang, Zhigang Lu, Haichen Wang +3
Differential privacy (DP) is the de facto privacy standard against privacy leakage attacks, including many recently discovered ones against large language models (LLMs). However, w…
Mitigating Overthinking in Large Reasoning Language Models via Reasoning Path Deviation Monitoring
Weixin Guan, Liang Li, Jiapeng Liu +6
Large Reasoning Language Models (LRLMs) demonstrate impressive capabilities on complex tasks by utilizing long Chain-of-Thought reasoning. However, they are prone to overthinking,…
Operational Semantics of Resolution and Productivity in Horn Clause Logic
Peng Fu, Ekaterina Komendantskaya
This paper presents a study of operational and type-theoretic properties of different resolution strategies in Horn clause logic. We distinguish four different kinds of resolution:…
DIVE into MoE: Diversity-Enhanced Reconstruction of Large Language Models from Dense into Mixture-of-Experts
Yuchen Feng, Bowen Shen, Naibin Gu +4
Large language models (LLMs) with the Mixture-of-Experts (MoE) architecture achieve high cost-efficiency by selectively activating a subset of the parameters. Despite the inference…
Near-Future Policy Optimization
Chuanyu Qin, Chenxu Yang, Qingyi Si +6
Reinforcement learning with verifiable rewards (RLVR) has become a core post-training recipe. Introducing suitable off-policy trajectories into on-policy exploration accelerates RL…
CBP-Tuning: Efficient Local Customization for Black-box Large Language Models
Jiaxuan Zhao, Naibin Gu, Yuchen Feng +4
The high costs of customizing large language models (LLMs) fundamentally limit their adaptability to user-specific needs. Consequently, LLMs are increasingly offered as cloud-based…
Are Large Language Models Table-based Fact-Checkers?
Hanwen Zhang, Qingyi Si, Peng Fu +2
Table-based Fact Verification (TFV) aims to extract the entailment relation between statements and structured tables. Existing TFV methods based on small-scaled models suffer from…
Linear Dependent Type Theory for Quantum Programming Languages
Peng Fu, Kohei Kishida, Peter Selinger
Modern quantum programming languages integrate quantum resources and classical control. They must, on the one hand, be linearly typed to reflect the no-cloning property of quantum…
SPOC learner's final grade prediction based on a novel sampling batch normalization embedded neural network method
Zhuonan Liang, Ziheng Liu, Huaze Shi +7
Recent years have witnessed the rapid growth of Small Private Online Courses (SPOC) which is able to highly customized and personalized to adapt variable educational requests, in w…
Learning to Win Lottery Tickets in BERT Transfer via Task-agnostic Mask Training
Yuanxin Liu, Fandong Meng, Zheng Lin +4
Recent studies on the lottery ticket hypothesis (LTH) show that pre-trained language models (PLMs) like BERT contain matching subnetworks that have similar transfer learning perfor…
Probing Phase Transition of Band Topology via Radiation Topology
Chang-Yin Ji, Wenze Lan, Peng Fu +5
Topological photonics has received extensive attention from researchers because it provides brand new physical principles to manipulate light. Band topology of optical materials is…
A Type-Theoretic Approach to Structural Resolution
Peng Fu, Ekaterina Komendantskaya
Structural resolution (or S-resolution) is a newly proposed alternative to SLD-resolution that allows a systematic separation of derivations into term-matching and unification step…
Hallucinations Leave a Grounding Signature:Verifier-Guided Decoding for Selective Object Correction
Lei Yang, Xinze Liu, Dayan Wu +7
The paper introduces a method to detect and correct hallucinated objects in large vision‑language models by identifying a hidden grounding pattern and using a lightweight verifier…
What to Format and How: A Benchmark and Workflow Approach for Document Formatting
Shihao Rao, Liang Li, Jiapeng Liu +6
Recent advances in large language models (LLMs) have opened up new possibilities for automated document formatting. However, real-world formatting often requires identifying target…
Co-Evolving Policy Distillation
Naibin Gu, Chenxu Yang, Qingyi Si +7
RLVR and OPD have become standard paradigms for post-training. We provide a unified analysis of these two paradigms in consolidating multiple expert capabilities into a single mode…
Proof Relevant Corecursive Resolution
Peng Fu, Ekaterina Komendantskaya, Tom Schrijvers +1
Resolution lies at the foundation of both logic programming and type class context reduction in functional languages. Terminating derivations by resolution have well-defined induct…
Visualization of Photonic Band Structures via Far-field Measurements in SiNx Photonic Crystal Slabs
Wenze Lan, Peng Fu, Chang-Yin Ji +4
The band structures of the photonic crystal slabs play a significant role in manipulating the flow of light and pre-dicting exotic physics in photonics. In this letter, we show tha…
Dependently Typed Folds for Nested Data Types
Peng Fu, Peter Selinger
We present an approach to develop folds for nested data types using dependent types. We call such folds , they have the following properties. (1)…
Learning to Solve, Forgetting to Retain: Correct-Set Turnover in RLVR
Chuanyu Qin, Chenxu Yang, Qingyi Si +3
Reinforcement learning with verifiable rewards (RLVR) improves the ability of large language model, yet headline accuracy gains often conceal a hidden cost: previously solved probl…
Revisiting the Knowledge Injection Frameworks
Peng Fu, Yiming Zhang, Haobo Wang +2
In recent years, large language models (LLMs), such as GPTs, have attained great impact worldwide. However, how to adapt these LLMs to better suit the vertical domain-specific task…
A Type Checking Algorithm for Higher-rank, Impredicative and Second-order Types
Peng Fu
We study a type checking algorithm that is able to type check a nontrivial subclass of functional programs that use features such as higher-rank, impredicative and second-order typ…
Language Prior Is Not the Only Shortcut: A Benchmark for Shortcut Learning in VQA
Qingyi Si, Fandong Meng, Mingyu Zheng +6
Visual Question Answering (VQA) models are prone to learn the shortcut solution formed by dataset biases rather than the intended solution. To evaluate the VQA models' reasoning ab…
A lateral semicircular canal segmentation based geometric calibration for human temporal bone CT Image
Xiaoguang Li, Peng Fu, Hongxia Yin +3
Computed Tomography (CT) of the temporal bone has become an important method for diagnosing ear diseases. Due to the different posture of the subject and the settings of CT scanner…
Advancing Academic Knowledge Retrieval via LLM-enhanced Representation Similarity Fusion
Wei Dai, Peng Fu, Chunjing Gan
In an era marked by robust technological growth and swift information renewal, furnishing researchers and the populace with top-tier, avant-garde academic insights spanning various…
A Hierarchical Transformer with Speaker Modeling for Emotion Recognition in Conversation
Jiangnan Li, Zheng Lin, Peng Fu +2
Emotion Recognition in Conversation (ERC) is a more challenging task than conventional text emotion recognition. It can be regarded as a personalized and interactive emotion recogn…
Learning Class-Transductive Intent Representations for Zero-shot Intent Detection
Qingyi Si, Yuanxin Liu, Peng Fu +3
Zero-shot intent detection (ZSID) aims to deal with the continuously emerging intents without annotated training data. However, existing ZSID systems suffer from two limitations: 1…
Proto-Quipper with Reversing and Control
Peng Fu, Kohei Kishida, Neil J. Ross +1
The quantum programming language Quipper supports circuit operations such as reversing and controlling certain quantum circuits. Additionally, Quipper provides a function called wi…
Multimodal Hypothetical Summary for Retrieval-based Multi-image Question Answering
Peize Li, Qingyi Si, Peng Fu +2
Retrieval-based multi-image question answering (QA) task involves retrieving multiple question-related images and synthesizing these images to generate an answer. Conventional "ret…
Towards an induction principle for nested data types
Peng Fu, Peter Selinger
A well-known problem in the theory of dependent types is how to handle so-called nested data types. These data types are difficult to program and to reason about in total dependent…
Beyond Post-Quantization: Native Hash Learning with a Dedicated HASH Token
Xinze Liu, Ding Wang, Hengjie Zhu +4
Efficient large-scale image retrieval requires compact representations that preserve semantic similarity under fast Hamming-space search. Deep hashing is appealing, but most existi…
Elastic MoE: Unlocking the Inference-Time Scalability of Mixture-of-Experts
Naibin Gu, Zhenyu Zhang, Yuchen Feng +8
Mixture-of-Experts (MoE) models typically fix the number of activated experts at both training and inference. However, real-world deployments often face heterogeneous hardware,…
Light-PEFT: Lightening Parameter-Efficient Fine-Tuning via Early Pruning
Naibin Gu, Peng Fu, Xiyu Liu +3
Parameter-efficient fine-tuning (PEFT) has emerged as the predominant technique for fine-tuning in the era of large language models. However, existing PEFT methods still have inade…
Check It Again: Progressive Visual Question Answering via Visual Entailment
Qingyi Si, Zheng Lin, Mingyu Zheng +2
While sophisticated Visual Question Answering models have achieved remarkable success, they tend to answer questions only according to superficial correlations between question and…
Towards Robust Visual Question Answering: Making the Most of Biased Samples via Contrastive Learning
Qingyi Si, Yuanxin Liu, Fandong Meng +5
Models for Visual Question Answering (VQA) often rely on the spurious correlations, i.e., the language priors, that appear in the biased samples of training set, which make them br…
Nonreciprocal quantum phase transition in cavity magnonics
Ye-jun Xu, Long-hua Zhai, Peng Fu +2
We investigate the nonreciprocal quantum phase transition in a cavity magnonic system driven by a parametric field, where an yttrium iron garnet (YIG) sphere is placed in a spinnin…
Proto-Quipper with dynamic lifting
Peng Fu, Kohei Kishida, Neil J. Ross +1
Quipper is a functional programming language for quantum computing. Proto-Quipper is a family of languages aiming to provide a formal foundation for Quipper. In this paper, we exte…
Question-Interlocutor Scope Realized Graph Modeling over Key Utterances for Dialogue Reading Comprehension
Jiangnan Li, Mo Yu, Fandong Meng +4
In this work, we focus on dialogue reading comprehension (DRC), a task extracting answer spans for questions from dialogues. Dialogue context modeling in DRC is tricky due to compl…
A Type-Theoretic Approach to Resolution
Peng Fu, Ekaterina Komendantskaya
We propose a new type-theoretic approach to SLD-resolution and Horn-clause logic programming. It views Horn formulas as types, and derivations for a given query as a construction o…
BeamLoRA: Beam-Constraint Low-Rank Adaptation
Naibin Gu, Zhenyu Zhang, Xiyu Liu +7
Due to the demand for efficient fine-tuning of large language models, Low-Rank Adaptation (LoRA) has been widely adopted as one of the most effective parameter-efficient fine-tunin…
Blink: Dynamic Visual Token Resolution for Enhanced Multimodal Understanding
Yuchen Feng, Zhenyu Zhang, Naibin Gu +8
Multimodal large language models (MLLMs) have achieved remarkable progress on various vision-language tasks, yet their visual perception remains limited. Humans, in comparison, per…