Search arXiv⌕ Search

arXiv subjects

Wei Wei

Publications and source records attributed to Wei Wei.

At least 163 records · Page 9Linked to original sources
arXiv2025

SFT-GO: Supervised Fine-Tuning with Group Optimization for Large Language Models

Supervised fine-tuning (SFT) has become an essential step in tailoring large language models (LLMs) to align with human expectations and specific downstream tasks. However, existing SFT methods typically treat each training instance as a uniform sequence, giving equal importance to all tokens regardless of their relevance. This overlooks the fact that only a subset of tokens often contains critical, task-specific information. To address this limitation, we introduce Supervised Fine-Tuning with Group Optimization (SFT-GO), a novel approach that treats groups of tokens differently based on their importance.SFT-GO groups tokens in each sample based on their importance values and optimizes the LLM using a weighted combination of the worst-group loss and the standard cross-entropy loss. This mechanism adaptively emphasizes the most challenging token groups and guides the model to better handle different group distributions, thereby improving overall learning dynamics. We provide a theoretical analysis of SFT-GO's convergence rate, demonstrating its efficiency. Empirically, we apply SFT-GO with three different token grouping strategies and show that models trained with SFT-GO consistently outperform baseline approaches across popular LLM benchmarks. These improvements hold across various datasets and base models, demonstrating the robustness and the effectiveness of our method.

cs.LG↗
arXiv2025

Merlin: Multi-View Representation Learning for Robust Multivariate Time Series Forecasting with Unfixed Missing Rates

Multivariate Time Series Forecasting (MTSF) involves predicting future values of multiple interrelated time series. Recently, deep learning-based MTSF models have gained significant attention for their promising ability to mine semantics (global and local information) within MTS data. However, these models are pervasively susceptible to missing values caused by malfunctioning data collectors. These missing values not only disrupt the semantics of MTS, but their distribution also changes over time. Nevertheless, existing models lack robustness to such issues, leading to suboptimal forecasting performance. To this end, in this paper, we propose Multi-View Representation Learning (Merlin), which can help existing models achieve semantic alignment between incomplete observations with different missing rates and complete observations in MTS. Specifically, Merlin consists of two key modules: offline knowledge distillation and multi-view contrastive learning. The former utilizes a teacher model to guide a student model in mining semantics from incomplete observations, similar to those obtainable from complete observations. The latter improves the student model's robustness by learning from positive/negative data pairs constructed from incomplete observations with different missing rates, ensuring semantic alignment across different missing rates. Therefore, Merlin is capable of effectively enhancing the robustness of existing models against unfixed missing rates while preserving forecasting accuracy. Experiments on four real-world datasets demonstrate the superiority of Merlin.

cs.LG↗
arXiv2025

Human-like object concept representations emerge naturally in multimodal large language models

Understanding how humans conceptualize and categorize natural objects offers critical insights into perception and cognition. With the advent of Large Language Models (LLMs), a key question arises: can these models develop human-like object representations from linguistic and multimodal data? In this study, we combined behavioral and neuroimaging analyses to explore the relationship between object concept representations in LLMs and human cognition. We collected 4.7 million triplet judgments from LLMs and Multimodal LLMs (MLLMs) to derive low-dimensional embeddings that capture the similarity structure of 1,854 natural objects. The resulting 66-dimensional embeddings were stable, predictive, and exhibited semantic clustering similar to human mental representations. Remarkably, the dimensions underlying these embeddings were interpretable, suggesting that LLMs and MLLMs develop human-like conceptual representations of objects. Further analysis showed strong alignment between model embeddings and neural activity patterns in brain regions such as EBA, PPA, RSC, and FFA. This provides compelling evidence that the object representations in LLMs, while not identical to human ones, share fundamental similarities that reflect key aspects of human conceptual knowledge. Our findings advance the understanding of machine intelligence and inform the development of more human-like artificial cognitive systems.

cs.AI↗
arXiv2025

Action-Adaptive Continual Learning: Enabling Policy Generalization under Dynamic Action Spaces

Continual Learning (CL) is a powerful tool that enables agents to learn a sequence of tasks, accumulating knowledge learned in the past and using it for problem-solving or future task learning. However, existing CL methods often assume that the agent's capabilities remain static within dynamic environments, which doesn't reflect real-world scenarios where capabilities dynamically change. This paper introduces a new and realistic problem: Continual Learning with Dynamic Capabilities (CL-DC), posing a significant challenge for CL agents: How can policy generalization across different action spaces be achieved? Inspired by the cortical functions, we propose an Action-Adaptive Continual Learning framework (AACL) to address this challenge. Our framework decouples the agent's policy from the specific action space by building an action representation space. For a new action space, the encoder-decoder of action representations is adaptively fine-tuned to maintain a balance between stability and plasticity. Furthermore, we release a benchmark based on three environments to validate the effectiveness of methods for CL-DC. Experimental results demonstrate that our framework outperforms popular methods by generalizing the policy across action spaces.

cs.LG↗
arXiv2025

Multi-granularity Knowledge Transfer for Continual Reinforcement Learning

Continual reinforcement learning (CRL) empowers RL agents with the ability to learn a sequence of tasks, accumulating knowledge learned in the past and using the knowledge for problemsolving or future task learning. However, existing methods often focus on transferring fine-grained knowledge across similar tasks, which neglects the multi-granularity structure of human cognitive control, resulting in insufficient knowledge transfer across diverse tasks. To enhance coarse-grained knowledge transfer, we propose a novel framework called MT-Core (as shorthand for Multi-granularity knowledge Transfer for Continual reinforcement learning). MT-Core has a key characteristic of multi-granularity policy learning: 1) a coarsegrained policy formulation for utilizing the powerful reasoning ability of the large language model (LLM) to set goals, and 2) a fine-grained policy learning through RL which is oriented by the goals. We also construct a new policy library (knowledge base) to store policies that can be retrieved for multi-granularity knowledge transfer. Experimental results demonstrate the superiority of the proposed MT-Core in handling diverse CRL tasks versus popular baselines.

cs.LG↗
arXiv2025

AdaptSFL: Adaptive Split Federated Learning in Resource-constrained Edge Networks

The increasing complexity of deep neural networks poses significant barriers to democratizing them to resource-limited edge devices. To address this challenge, split federated learning (SFL) has emerged as a promising solution by of floading the primary training workload to a server via model partitioning while enabling parallel training among edge devices. However, although system optimization substantially influences the performance of SFL under resource-constrained systems, the problem remains largely uncharted. In this paper, we provide a convergence analysis of SFL which quantifies the impact of model splitting (MS) and client-side model aggregation (MA) on the learning performance, serving as a theoretical foundation. Then, we propose AdaptSFL, a novel resource-adaptive SFL framework, to expedite SFL under resource-constrained edge computing systems. Specifically, AdaptSFL adaptively controls client-side MA and MS to balance communication-computing latency and training convergence. Extensive simulations across various datasets validate that our proposed AdaptSFL framework takes considerably less time to achieve a target accuracy than benchmarks, demonstrating the effectiveness of the proposed strategies.

cs.LG↗
arXiv2025

VecTrans: Enhancing Compiler Auto-Vectorization through LLM-Assisted Code Transformations

Auto-vectorization is a fundamental optimization for modern compilers to exploit SIMD parallelism. However, state-of-the-art approaches still struggle to handle intricate code patterns, often requiring manual hints or domain-specific expertise. Large language models (LLMs), with their ability to capture intricate patterns, provide a promising solution, yet their effective application in compiler optimizations remains an open challenge due to issues such as hallucinations and a lack of domain-specific reasoning. In this paper, we present VecTrans, a novel framework that leverages LLMs to enhance compiler-based code vectorization. VecTrans first employs compiler analysis to identify potentially vectorizable code regions. It then utilizes an LLM to refactor these regions into patterns that are more amenable to the compilers auto-vectorization. To ensure semantic correctness, VecTrans further integrates a hybrid validation mechanism at the intermediate representation (IR) level. With the above efforts, VecTrans combines the adaptability of LLMs with the precision of compiler vectorization, thereby effectively opening up the vectorization opportunities. experimental results show that among all TSVC functions unvectorizable by GCC, ICC, Clang, and BiSheng Compiler, VecTrans achieves an geomean speedup of 1.77x and successfully vectorizes 24 of 51 test cases. This marks a significant advancement over state-of-the-art approaches while maintaining a cost efficiency of $0.012 per function optimization for LLM API usage.

cs.SE↗
arXiv2025

MMSciBench: Benchmarking Language Models on Chinese Multimodal Scientific Problems

Recent advances in large language models (LLMs) and vision-language models (LVLMs) have shown promise across many tasks, yet their scientific reasoning capabilities remain untested, particularly in multimodal settings. We present MMSciBench, a benchmark for evaluating mathematical and physical reasoning through text-only and text-image formats, with human-annotated difficulty levels, solutions with detailed explanations, and taxonomic mappings. Evaluation of state-of-the-art models reveals significant limitations, with even the best model achieving only \textbf{63.77\%} accuracy and particularly struggling with visual reasoning tasks. Our analysis exposes critical gaps in complex reasoning and visual-textual integration, establishing MMSciBench as a rigorous standard for measuring progress in multimodal scientific understanding. The code for MMSciBench is open-sourced at GitHub, and the dataset is available at Hugging Face.

cs.LG↗
arXiv2025

Risk-aware Direct Preference Optimization under Nested Risk Measure

When fine-tuning pre-trained Large Language Models (LLMs) to align with human values and intentions, maximizing the estimated reward can lead to superior performance, but it also introduces potential risks due to deviations from the reference model's intended behavior. Most existing methods typically introduce KL divergence to constrain deviations between the trained model and the reference model; however, this may not be sufficient in certain applications that require tight risk control. In this paper, we introduce Risk-aware Direct Preference Optimization (Ra-DPO), a novel approach that incorporates risk-awareness by employing a class of nested risk measures. This approach formulates a constrained risk-aware advantage function maximization problem and then converts the Bradley-Terry model into a token-level representation. The objective function maximizes the likelihood of the policy while suppressing the deviation between a trained model and the reference model using a sequential risk ratio, thereby enhancing the model's risk-awareness. Experimental results across three open-source datasets: IMDb Dataset, Anthropic HH Dataset, and AlpacaEval, demonstrate the proposed method's superior performance in balancing alignment performance and model drift. Our code is opensourced at https://github.com/zlj123-max/Ra-DPO.

cs.LG↗
arXiv2025

Enhancing Retrieval for ESGLLM via ESG-CID -- A Disclosure Content Index Finetuning Dataset for Mapping GRI and ESRS

Climate change has intensified the need for transparency and accountability in organizational practices, making Environmental, Social, and Governance (ESG) reporting increasingly crucial. Frameworks like the Global Reporting Initiative (GRI) and the new European Sustainability Reporting Standards (ESRS) aim to standardize ESG reporting, yet generating comprehensive reports remains challenging due to the considerable length of ESG documents and variability in company reporting styles. To facilitate ESG report automation, Retrieval-Augmented Generation (RAG) systems can be employed, but their development is hindered by a lack of labeled data suitable for training retrieval models. In this paper, we leverage an underutilized source of weak supervision -- the disclosure content index found in past ESG reports -- to create a comprehensive dataset, ESG-CID, for both GRI and ESRS standards. By extracting mappings between specific disclosure requirements and corresponding report sections, and refining them using a Large Language Model as a judge, we generate a robust training and evaluation set. We benchmark popular embedding models on this dataset and show that fine-tuning BERT-based models can outperform commercial embeddings and leading public models, even under temporal data splits for cross-report style transfer from GRI to ESRS. Data: https://huggingface.co/datasets/airefinery/esg_cid_retrieval

cs.CL↗
arXiv2025

Collaborative Memory: Multi-User Memory Sharing in LLM Agents with Dynamic Access Control

Complex tasks are increasingly delegated to ensembles of specialized LLM-based agents that reason, communicate, and coordinate actions-both among themselves and through interactions with external tools, APIs, and databases. While persistent memory has been shown to enhance single-agent performance, most approaches assume a monolithic, single-user context-overlooking the benefits and challenges of knowledge transfer across users under dynamic, asymmetric permissions. We introduce Collaborative Memory, a framework for multi-user, multi-agent environments with asymmetric, time-evolving access controls encoded as bipartite graphs linking users, agents, and resources. Our system maintains two memory tiers: (1) private memory-private fragments visible only to their originating user; and (2) shared memory-selectively shared fragments. Each fragment carries immutable provenance attributes (contributing agents, accessed resources, and timestamps) to support retrospective permission checks. Granular read policies enforce current user-agent-resource constraints and project existing memory fragments into filtered transformed views. Write policies determine fragment retention and sharing, applying context-aware transformations to update the memory. Both policies may be designed conditioned on system, agent, and user-level information. Our framework enables safe, efficient, and interpretable cross-user knowledge sharing, with provable adherence to asymmetric, time-varying policies and full auditability of memory operations.

cs.MA↗
arXiv2025

Solutions to Equilibrium HJB Equations for Time-Inconsistent Deterministic Linear Quadratic Control: Characterization and Uniqueness

In this paper we study a class of HJB equations which solve for equilibria for general time-inconsistent deterministic linear quadratic control problems within the intra-personal game theoretic framework, where the inconsistency arises from non-exponential discount functions. We characterize the solutions to the HJB equations using a class of Riccati equations with integral terms. By studying the uniqueness of solutions to the integro-differential Riccati equations, we prove the uniqueness of solutions to the equilibrium HJB equations.

math.OC↗
arXiv2025

Recursive Optimal Stopping with Poisson Stopping Constraints

This paper solves a recursive optimal stopping problem with Poisson stopping constraints using the penalized backward stochastic differential equation (PBSDE) with jumps. Stopping in this problem is only allowed at Poisson random intervention times, and jumps play a significant role not only through the stopping times but also in the recursive objective functional and model coefficients. To solve the problem, we propose a decomposition method based on Jacod-Pham that allows us to separate the problem into a series of sub-problems between each pair of consecutive Poisson stopping times. To represent the value function of the recursive optimal stopping problem when the initial time falls between two consecutive Poisson stopping times and the generator is concave/convex, we leverage the comparison theorem of BSDEs with jumps. We then apply the representation result to American option pricing in a nonlinear market with Poisson stopping constraints.

math.OC↗
arXiv2025

Enhanced Sample Selection with Confidence Tracking: Identifying Correctly Labeled yet Hard-to-Learn Samples in Noisy Data

We propose a novel sample selection method for image classification in the presence of noisy labels. Existing methods typically consider small-loss samples as correctly labeled. However, some correctly labeled samples are inherently difficult for the model to learn and can exhibit high loss similar to mislabeled samples in the early stages of training. Consequently, setting a threshold on per-sample loss to select correct labels results in a trade-off between precision and recall in sample selection: a lower threshold may miss many correctly labeled hard-to-learn samples (low recall), while a higher threshold may include many mislabeled samples (low precision). To address this issue, our goal is to accurately distinguish correctly labeled yet hard-to-learn samples from mislabeled ones, thus alleviating the trade-off dilemma. We achieve this by considering the trends in model prediction confidence rather than relying solely on loss values. Empirical observations show that only for correctly labeled samples, the model's prediction confidence for the annotated labels typically increases faster than for any other classes. Based on this insight, we propose tracking the confidence gaps between the annotated labels and other classes during training and evaluating their trends using the Mann-Kendall Test. A sample is considered potentially correctly labeled if all its confidence gaps tend to increase. Our method functions as a plug-and-play component that can be seamlessly integrated into existing sample selection techniques. Experiments on several standard benchmarks and real-world datasets demonstrate that our method enhances the performance of existing methods for learning with noisy labels.

cs.CV↗
arXiv2025

Hierarchical Split Federated Learning: Convergence Analysis and System Optimization

As AI models expand in size, it has become increasingly challenging to deploy federated learning (FL) on resource-constrained edge devices. To tackle this issue, split federated learning (SFL) has emerged as an FL framework with reduced workload on edge devices via model splitting; it has received extensive attention from the research community in recent years. Nevertheless, most prior works on SFL focus only on a two-tier architecture without harnessing multi-tier cloudedge computing resources. In this paper, we intend to analyze and optimize the learning performance of SFL under multi-tier systems. Specifically, we propose the hierarchical SFL (HSFL) framework and derive its convergence bound. Based on the theoretical results, we formulate a joint optimization problem for model splitting (MS) and model aggregation (MA). To solve this rather hard problem, we then decompose it into MS and MA subproblems that can be solved via an iterative descending algorithm. Simulation results demonstrate that the tailored algorithm can effectively optimize MS and MA for SFL within virtually any multi-tier system.

cs.LG↗
arXiv2025

Learning a Canonical Basis of Human Preferences from Binary Ratings

Recent advances in generative AI have been driven by alignment techniques such as reinforcement learning from human feedback (RLHF). RLHF and related techniques typically involve constructing a dataset of binary or ranked choice human preferences and subsequently fine-tuning models to align with these preferences. This paper shifts the focus to understanding the preferences encoded in such datasets and identifying common human preferences. We find that a small subset of 21 preference categories (selected from a set of nearly 5,000 distinct preferences) captures >89% of preference variation across individuals. This small set of preferences is analogous to a canonical basis of human preferences, similar to established findings that characterize human variation in psychology or facial recognition studies. Through both synthetic and empirical evaluations, we confirm that our low-rank, canonical set of human preferences generalizes across the entire dataset and within specific topics. We further demonstrate our preference basis' utility in model evaluation, where our preference categories offer deeper insights into model alignment, and in model training, where we show that fine-tuning on preference-defined subsets successfully aligns the model accordingly.

cs.LG↗
arXiv2025

Advances in Continual Graph Learning for Anti-Money Laundering Systems: A Comprehensive Review

Financial institutions are required by regulation to report suspicious financial transactions related to money laundering. Therefore, they need to constantly monitor vast amounts of incoming and outgoing transactions. A particular challenge in detecting money laundering is that money launderers continuously adapt their tactics to evade detection. Hence, detection methods need constant fine-tuning. Traditional machine learning models suffer from catastrophic forgetting when fine-tuning the model on new data, thereby limiting their effectiveness in dynamic environments. Continual learning methods may address this issue and enhance current anti-money laundering (AML) practices, by allowing models to incorporate new information while retaining prior knowledge. Research on continual graph learning for AML, however, is still scarce. In this review, we critically evaluate state-of-the-art continual graph learning approaches for AML applications. We categorise methods into replay-based, regularization-based, and architecture-based strategies within the graph neural network (GNN) framework, and we provide in-depth experimental evaluations on both synthetic and real-world AML data sets that showcase the effect of the different hyperparameters. Our analysis demonstrates that continual learning improves model adaptability and robustness in the face of extreme class imbalances and evolving fraud patterns. Finally, we outline key challenges and propose directions for future research.

cs.LG↗