Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Kun ZhangSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works39 from public data
- Effective Neural Topic Modeling with Embedding Clustering Regularization99
A new neural topic model is proposed, Embedding Clustering Regularization Topic Model (ECRTM), which forces each topic embedding to be the center of a separately aggregated word embedding cluster in the semantic space, which alleviates topic collapsing.
- 61
- Certified Robustness Against Natural Language Attacks by Causal Intervention43
Causal Intervention by Semantic Smoothing (CISS), a novel framework towards robustness against natural language attacks, learns causal effects p(y|do(x) by smoothing in the latent semantic space to make robust predictions, which scales to deep architectures and avoids tedious construction of noise customized for specific attacks.
- A Versatile Causal Discovery Framework to Allow Causally-Related Hidden Variables38
A novel, versatile framework for causal discovery that accommodates the presence of causally-related hidden variables almost everywhere in the causal network, based on rank information of covariance matrix over observed variables is introduced.
- 38
- Score-Based Causal Discovery of Latent Variable Causal Models24
This work develops score-based methods that are capable of identifying causal structures containing causally-related latent variables with identifiability guarantees and shows that a properly formulated scoring function can achieve score equivalence and consistency for structure learning of latent variable causal models.
- Topic Modeling as Multi-Objective Contrastive Optimization16
A novel contrastive learning method oriented towards sets of topic vectors to capture useful semantics that are shared among a set of input documents is introduced, with the goal of achieving a Pareto stationary solution that balances the trade-off between the ELBO and the contrastive objective.
- Causal Representation Learning from General Environments under Nonparametric Mixing14
It is shown that one can fully recover the latent DAG and identify the latent variables up to minor indeterminacies under a nonparametric mixing function and nonlinear latent causal models, such as additive (Gaussian) noise models or heteroscedastic noise models, by properly leveraging sufficient change conditions on the causal mechanisms up to third-order derivatives.
- MAMA: Meta-optimized Angular Margin Contrastive Framework for Video-Language Representation Learning12
To adapt to the non-uniform concept distribution, MAMA utilizes a multi-layer perceptron (MLP)-parameterized weighting function that maps loss values to sample weights which enable dynamic adjustment of the model's focus throughout the training.
- 12
- DemaFormer: Damped Exponential Moving Average Transformer with Energy-Based Modeling for Temporal Language Grounding12
An energy-based model framework to explicitly learn moment- query distributions is proposed and DemaFormer, a novel Transformer-based architecture that utilizes exponential moving average with a learnable damping factor to effectively encode moment-query inputs is proposed.
- 9
- API-Net: Robust Generative Classifier via a Single Discriminator9
An Anti-Perturbation Inference (API) method, which searches for anti-perturbations to maximize the lower bound of the joint log-likelihood of inputs and classes, and is constructed upon a generative classifiers upon a single discriminator, which takes advantage of the generative properties to tackle off-manifold examples while maintaining a succinct structure for effective optimization.
- Advancing Reasoning in Diffusion Language Models with Denoising Process Rewards8
This work introduces a process-level reinforcement signal defined over the denoising trajectory of diffusion language models, and proposes an efficient stochastic estimator that reuses standard training rollouts, enabling practical process-level supervision at scale.
- Synergy Between Sufficient Changes and Sparse Mixing Procedure for Disentangled Representation Learning6
An identifiability theory with less restrictive constraints regarding distribution changes and the sparse mixing procedure is proposed, enhancing applicability to real-world scenarios and developing an estimation framework incorporating a domain encoding network and a sparse mixing constraint.
- 6
- Permutation-Based Rank Test in the Presence of Discretization and Application in Causal Discovery with Mixed Data5
MPRT can effectively control the Type I error in the presence of discretization while previous methods cannot, and is validated by extensive experiments on synthetic data and real-world data to demonstrate its effectiveness as well as applicability in causal discovery.
- Generative Framework for Personalized Persuasion: Inferring Causal, Counterfactual, and Latent Knowledge5
This work considers the psychological constructs and unobservable noises that might be influencing user-system interactions as latent factors and model the user utterance strategies as causal factors, enabling system strategies to be treated as counterfactual actions.
- Causal discovery and counterfactual reasoning to optimize persuasive dialogue policies5
A novel method that leverages causal discovery and counterfactual reasoning for optimising system persuasion capability and outcomes and optimising reinforcement learning policies for online dialogue systems is presented.
- READ-PVLA: Recurrent Adapter with Partial Video-Language Alignment for Parameter-Efficient Transfer Learning in Low-Resource Video-Language Modeling4
A novel REcurrent ADapter (READ) that employs recurrent computation to enable temporal modeling capability and Partial Video-Language Alignment (PVLA) objective via the use of partial optimal transport to maintain task-related information flowing into the authors' READ modules is proposed.
- Counterfactual Reasoning Using Predicted Latent Personality Dimensions for Optimizing Persuasion Outcome4
A novel approach that tracks a user's latent personality dimensions (LPDs) during ongoing persuasion conversation and generates tailored counterfactual utterances based on these LPDs to optimize the overall persuasion outcome is presented.
- 4
- Score-based Greedy Search for Structure Identification of Partially Observed Linear Causal Models3
This work proposes the first score-based greedy search method for the identification of structure involving latent variables with identifiability guarantees and designs Latent variable Greedy Equivalence Search, a greedy search algorithm for this class of model with well-defined operators, which search very efficiently over the graph space to find the optimal structure.
- 3
- A Sample Efficient Conditional Independence Test in the Presence of Discretization2
This paper finds that the independence relationships of latent continuous variables can be established by addressing an over-identifying restriction problem with Generalized Method of Moments (GMM), and derives an appropriate test statistic and establishes its asymptotic distribution correctly reflecting CI by leveraging nodewise regression.
- Identifying Latent State-Transition Processes for Individualized Reinforcement Learning2
Experiments show that the proposed method can effectively identify latent state-transition processes and facilitate the learning of individualized RL policies.
- The Power of Order: Fooling LLMs with Adversarial Table Permutations1
This paper demonstrates that modern LLMs exhibit a significant vulnerability to the layout of tabular data, and introduces Adversarial Table Permutation, a novel, gradient-based attack that efficiently identifies worst-case permutations designed to maximally disrupt model performance.
- Representation Interventions Enable Lifelong Knowledge Memory Control in LLMs1
This work introduces RILKE (Representation Intervention for Lifelong KnowledgE Control), a robust and scalable method that treats knowledge control as interventions within the model's representation space and identifies two key properties enabling RILKE to achieve fine-grained control over complex, unstructured knowledge while maintaining general utility with frozen base weights.
- 1
- Beyond Perplexity: A Behavioral Evaluation Framework for Deployment-Memory Claims in LLM Test-Time Training–
A behavioral evaluation framework that calibrates TTT memory claims to the evidence that supports them and gives authors and evaluators a concrete standard for aligning TTT memory claims with the evidence actually reported is introduced.
- –
- –
- –
- Type Information-Assisted Self-Supervised Knowledge Graph Denoising–
This paper proposes a novel self-supervised knowledge graph denoising method that exploits the consistency between entity and relation type information for noise detection, resulting a novel self-supervised knowledge graph denoising method that avoids problems of knowledge graph compression and completion.
- –
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.