Khiem Le

Giang Do, Khiem Le, Quang Pham et al.

By routing input tokens to only a few split experts, Sparse Mixture-of-Experts has enabled efficient training of large language models. Recent findings suggest that fixing the routers can achieve competitive performance by alleviating the collapsing problem, where all experts eventually learn similar representations. However, this strategy has two key limitations: (i) the policy derived from random routers might be sub-optimal, and (ii) it requires extensive resources during training and evaluation, leading to limited efficiency gains. This work introduces \HyperRout, which dynamically generates the router's parameters through a fixed hypernetwork and trainable embeddings to achieve a balance between training the routers and freezing them to learn an improved routing policy. Extensive experiments across a wide range of tasks demonstrate the superior performance and efficiency gains of \HyperRouter compared to existing routing methods. Our implementation is publicly available at {\url{https://github.com/giangdip2410/HyperRouter}}.

Khiem Le, Youssef Mroueh, Phuc Nguyen et al.

Large language models trained via next-token prediction are fundamentally pattern-matchers: sensitive to superficial phrasing variations even when the underlying problem is identical. Group Relative Policy Optimization (GRPO) was designed to improve reasoning, but in fact it worsens this situation through two failure modes: diversity collapse, where training amplifies a single solution strategy while ignoring alternatives of gradient signal, and gradient diminishing, where a large portion of questions yield zero gradients because all rollouts receive identical rewards. We propose TA-GRPO (Transform-Augmented GRPO), which generates semantically equivalent transformed variants of each question (via paraphrasing, variable renaming, and format changes) and computes advantages by pooling rewards across the entire group. This pooled computation ensures mixed rewards even when the original question is too easy or too hard, while training on diverse phrasings promotes multiple solution strategies. We provide theoretical justification showing that TA-GRPO reduces zero-gradient probability and improves generalization via reduced train-test distribution shift. Experiments on mathematical reasoning benchmarks show consistent Pass@k improvements, with gains up to 9.84 points on competition math (AMC12, AIME24) and 5.05 points on out-of-distribution scientific reasoning (GPQA-Diamond).

Khiem Le, Long Ho, Cuong Do et al.

Domain shift is a formidable issue in Machine Learning that causes a model to suffer from performance degradation when tested on unseen domains. Federated Domain Generalization (FedDG) attempts to train a global model using collaborative clients in a privacy-preserving manner that can generalize well to unseen clients possibly with domain shift. However, most existing FedDG methods either cause additional privacy risks of data leakage or induce significant costs in client communication and computation, which are major concerns in the Federated Learning paradigm. To circumvent these challenges, here we introduce a novel architectural method for FedDG, namely gPerXAN, which relies on a normalization scheme working with a guiding regularizer. In particular, we carefully design Personalized eXplicitly Assembled Normalization to enforce client models selectively filtering domain-specific features that are biased towards local data while retaining discrimination of those features. Then, we incorporate a simple yet effective regularizer to guide these models in directly capturing domain-invariant representations that the global model's classifier can leverage. Extensive experimental results on two benchmark datasets, i.e., PACS and Office-Home, and a real-world medical dataset, Camelyon17, indicate that our proposed method outperforms other existing methods in addressing this particular problem.

Khiem Le, Ting Hua, Nitesh V. Chawla

Molecular optimization -- modifying a given molecule to improve desired properties -- is a fundamental task in drug discovery. While LLMs hold the potential to solve this task using natural language to drive the optimization, straightforward prompting achieves limited accuracy. In this work, we propose AgentDrug, an agentic workflow that leverages LLMs in a structured refinement process to achieve significantly higher accuracy. AgentDrug defines a nested refinement loop: the inner loop uses feedback from cheminformatics toolkits to validate molecular structures, while the outer loop guides the LLM with generic feedback and a gradient-based objective to steer the molecule toward property improvement. We evaluate AgentDrug on benchmarks with both single- and multi-property optimization under loose and strict thresholds. Results demonstrate significant performance gains over previous methods. With Qwen-2.5-3B, AgentDrug improves accuracy by 20.7\% (loose) and 16.8\% (strict) on six single-property tasks, and by 7.0\% and 5.3\% on eight multi-property tasks. With larger model Qwen-2.5-7B, AgentDrug further improves accuracy on 6 single-property objectives by 28.9\% (loose) and 29.0\% (strict), and on 8 multi-property objectives by 14.9\% (loose) and 13.2\% (strict).

Khiem Le, Sreejata Dey, Marcos Martínez Galindo et al.

Molecular Property Prediction (MPP) is a central task in drug discovery. While Large Language Models (LLMs) show promise as generalist models for MPP, their current performance remains below the threshold for practical adoption. We propose TreeKD, a novel knowledge distillation method that transfers complementary knowledge from tree-based specialist models into LLMs. Our approach trains specialist decision trees on functional group features, then verbalizes their learned predictive rules as natural language to enable rule-augmented context learning. This enables LLMs to leverage structural insights that are difficult to extract from SMILES strings alone. We further introduce rule-consistency, a test-time scaling technique inspired by bagging that ensembles predictions across diverse rules from a Random Forest. Experiments on 22 ADMET properties from the TDC benchmark demonstrate that TreeKD substantially improves LLM performance, narrowing the gap with SOTA specialist models and advancing toward practical generalist models for molecular property prediction.

Khiem Le, Tuan Tran, Ting Hua et al.

Existing resource-adaptive LoRA federated fine-tuning methods enable clients to fine-tune models using compressed versions of global LoRA matrices, in order to accommodate various compute resources across clients. This compression requirement will lead to suboptimal performance due to information loss. To address this, we propose FLAME, a novel federated learning framework based on the Sparse Mixture-of-Experts (SMoE) architecture. Unlike prior approaches, FLAME retains full (uncompressed) global LoRA matrices and achieves client-side adaptability by varying the number of activated experts per client. However, incorporating SMoE into federated learning introduces unique challenges, specifically, the mismatch in output magnitude from partial expert activation and the imbalance in expert training quality across clients. FLAME tackles these challenges through a lightweight rescaling mechanism and an activation-aware aggregation scheme. Empirical results across diverse computational settings demonstrate that FLAME consistently outperforms existing methods, providing a robust and effective solution for resource-adaptive federated learning.

Khiem Le, Zhichun Guo, Kaiwen Dong et al.

Large Language Models (LLMs) with their strong task-handling capabilities have shown remarkable advancements across a spectrum of fields, moving beyond natural language understanding. However, their proficiency within the chemistry domain remains restricted, especially in solving molecule-related tasks. This challenge is attributed to their inherent limitations in comprehending molecules using only common textual representations, i.e. SMILES strings. In this study, we seek to enhance the ability of LLMs to comprehend molecules by equipping them with a multi-modal external module, termed MolX. Instead of directly using SMILES strings to represent a molecule, we utilize specific encoders to extract fine-grained features from both SMILES string and 2D molecular graph representations for feeding into an LLM. A hand-crafted molecular fingerprint is incorporated to leverage its embedded domain knowledge. To establish an alignment between MolX and the LLM's textual input space, the model in which the LLM is frozen, is pre-trained with a strategy including a diverse set of tasks. Experimental evaluations show that our proposed method outperforms baselines across downstream molecule-related tasks ranging from molecule-to-text translation to molecular property prediction, with and without fine-tuning the LLM, while only introducing a small number of trainable parameters-0.53% and 0.82%, respectively.