Tootfinder

This https://arxiv.org/abs/2505.19912 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

APE: Selective Fine-tuning with Acceptance Criteria for Language Model Adaptation
We present Adjacent Possible Exploration (APE), a selective fine-tuning method for adapting large language models that systematically explores parameter modifications while maintaining model stability. Inspired by evolutionary optimization principles, APE evaluates multiple candidate parameter updates through fine-tuning on small data subsets and accepts only those exceeding a performance threshold. Unlike standard fine-tuning that follows single gradient directions, APE implements a filtered s…

TGRPO :Fine-tuning Vision-Language-Action Model via Trajectory-wise Group Relative Policy Optimization
Zengjue Chen, Runliang Niu, He Kong, Qi Wang
https://arxiv.org/abs/2506.08440

TGRPO :Fine-tuning Vision-Language-Action Model via Trajectory-wise Group Relative Policy Optimization
Recent advances in Vision-Language-Action (VLA) model have demonstrated strong generalization capabilities across diverse scenes, tasks, and robotic platforms when pretrained at large-scale datasets. However, these models still require task-specific fine-tuning in novel environments, a process that relies almost exclusively on supervised fine-tuning (SFT) using static trajectory datasets. Such approaches neither allow robot to interact with environment nor do they leverage feedback from live ex…

May I have your Attention? Breaking Fine-Tuning based Prompt Injection Defenses using Architecture-Aware Attacks
Nishit V. Pandya, Andrey Labunets, Sicun Gao, Earlence Fernandes
https://arxiv.org/abs/2507.07417

May I have your Attention? Breaking Fine-Tuning based Prompt Injection Defenses using Architecture-Aware Attacks
A popular class of defenses against prompt injection attacks on large language models (LLMs) relies on fine-tuning the model to separate instructions and data, so that the LLM does not follow instructions that might be present with data. There are several academic systems and production-level implementations of this idea. We evaluate the robustness of this class of prompt injection defenses in the whitebox setting by constructing strong optimization-based attacks and showing that the defenses d…

A Study on the Fine-Tuning Performance of Universal Machine-Learned Interatomic Potentials (U-MLIPs)
Xiaoqing Liu, Kehan Zeng, Yangshuai Wang, Teng Zhao
https://arxiv.org/abs/2506.07401

A Study on the Fine-Tuning Performance of Universal Machine-Learned Interatomic Potentials (U-MLIPs)
Universal machine-learned interatomic potentials (U-MLIPs) have demonstrated effectiveness across diverse atomistic systems but often require fine-tuning for task-specific accuracy. We investigate the fine-tuning of two MACE-based foundation models, MACE-MP-0 and its variant MACE-MP-0b, and identify key insights. Fine-tuning on task-specific datasets enhances accuracy and, in some cases, outperforms models trained from scratch. Additionally, fine-tuned models benefit from faster convergence due…

This https://arxiv.org/abs/2506.02308 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

MINT: Multimodal Instruction Tuning with Multimodal Interaction Grouping
Recent advances in multimodal foundation models have achieved state-of-the-art performance across a range of tasks. These breakthroughs are largely driven by new pre-training paradigms that leverage large-scale, unlabeled multimodal data, followed by instruction fine-tuning on curated labeled datasets and high-quality prompts. While there is growing interest in scaling instruction fine-tuning to ever-larger datasets in both quantity and scale, our findings reveal that simply increasing the numb…

DogFit: Domain-guided Fine-tuning for Efficient Transfer Learning of Diffusion Models
Yara Bahram, Mohammadhadi Shateri, Eric Granger
https://arxiv.org/abs/2508.05685 https://…

DogFit: Domain-guided Fine-tuning for Efficient Transfer Learning of Diffusion Models
Transfer learning of diffusion models to smaller target domains is challenging, as naively fine-tuning the model often results in poor generalization. Test-time guidance methods help mitigate this by offering controllable improvements in image fidelity through a trade-off with sample diversity. However, this benefit comes at a high computational cost, typically requiring dual forward passes during sampling. We propose the Domain-guided Fine-tuning (DogFit) method, an effective guidance mechanis…

Reinforcement Fine-Tuning for Reasoning towards Multi-Step Multi-Source Search in Large Language Models
Wentao Shi, Yiqing Shen
https://arxiv.org/abs/2506.08352

Reinforcement Fine-Tuning for Reasoning towards Multi-Step Multi-Source Search in Large Language Models
Large language models (LLMs) can face factual limitations when responding to time-sensitive queries about recent events that arise after their knowledge thresholds in the training corpus. Existing search-augmented approaches fall into two categories, each with distinct limitations: multi-agent search frameworks incur substantial computational overhead by separating search planning and response synthesis across multiple LLMs, while single-LLM tool-calling methods restrict themselves to sequentia…

Symbiosis: Multi-Adapter Inference and Fine-Tuning
Saransh Gupta, Umesh Deshpande, Travis Janssen, Swami Sundararaman
https://arxiv.org/abs/2507.03220 http…

Symbiosis: Multi-Adapter Inference and Fine-Tuning
Parameter-efficient fine-tuning (PEFT) allows model builders to capture the task specific parameters into adapters, which are a fraction of the size of the original base model. Popularity of PEFT technique for fine-tuning has led to creation of a large number of adapters for popular Large Language Models (LLMs). However, existing frameworks fall short in supporting inference or fine-tuning with multiple adapters in the following ways. 1) For fine-tuning, each job needs to deploy its dedicated b…

Narrowing the Gap: Supervised Fine-Tuning of Open-Source LLMs as a Viable Alternative to Proprietary Models for Pedagogical Tools
Lorenzo Lee Solano, Charles Koutcheme, Juho Leinonen, Alexandra Vassar, Jake Renzella
https://arxiv.org/abs/2507.05305

Narrowing the Gap: Supervised Fine-Tuning of Open-Source LLMs as a Viable Alternative to Proprietary Models for Pedagogical Tools
Frontier Large language models (LLMs) like ChatGPT and Gemini can decipher cryptic compiler errors for novice programmers, but their computational scale, cost, and tendency to over-assist make them problematic for widespread pedagogical adoption. This work demonstrates that smaller, specialised language models, enhanced via Supervised Fine-Tuning (SFT), present a more viable alternative for educational tools. We utilise a new dataset of 40,000 C compiler error explanations, derived from real in…

DMFI: Dual-Modality Fine-Tuning and Inference Framework for LLM-Based Insider Threat Detection
Kaichuan Kong, Dongjie Liu, Xiaobo Jin, Guanggang Geng, Zhiying Li, Jian Weng
https://arxiv.org/abs/2508.05694

DMFI: Dual-Modality Fine-Tuning and Inference Framework for LLM-Based Insider Threat Detection
Insider threat detection (ITD) poses a persistent and high-impact challenge in cybersecurity due to the subtle, long-term, and context-dependent nature of malicious insider behaviors. Traditional models often struggle to capture semantic intent and complex behavior dynamics, while existing LLM-based solutions face limitations in prompt adaptability and modality coverage. To bridge this gap, we propose DMFI, a dual-modality framework that integrates semantic inference with behavior-aware fine-tu…

Mitigating Catastrophic Forgetting with Adaptive Transformer Block Expansion in Federated Fine-Tuning
Yujia Huo, Jianchun Liu, Hongli Xu, Zhenguo Ma, Shilong Wang, Liusheng Huang
https://arxiv.org/abs/2506.05977

Mitigating Catastrophic Forgetting with Adaptive Transformer Block Expansion in Federated Fine-Tuning
Federated fine-tuning (FedFT) of large language models (LLMs) has emerged as a promising solution for adapting models to distributed data environments while ensuring data privacy. Existing FedFT methods predominantly utilize parameter-efficient fine-tuning (PEFT) techniques to reduce communication and computation overhead. However, they often fail to adequately address the catastrophic forgetting, a critical challenge arising from continual adaptation in distributed environments. The tradit…

KeyKnowledgeRAG (K^2RAG): An Enhanced RAG method for improved LLM question-answering capabilities
Hruday Markondapatnaikuni, Basem Suleiman, Abdelkarim Erradi, Shijing Chen
https://arxiv.org/abs/2507.07695

KeyKnowledgeRAG (K^2RAG): An Enhanced RAG method for improved LLM question-answering capabilities
Fine-tuning is an immensely resource-intensive process when retraining Large Language Models (LLMs) to incorporate a larger body of knowledge. Although many fine-tuning techniques have been developed to reduce the time and computational cost involved, the challenge persists as LLMs continue to grow in size and complexity. To address this, a new approach to knowledge expansion in LLMs is needed. Retrieval-Augmented Generation (RAG) offers one such alternative by storing external knowledge in a d…

Video-RTS: Rethinking Reinforcement Learning and Test-Time Scaling for Efficient and Enhanced Video Reasoning
Ziyang Wang, Jaehong Yoon, Shoubin Yu, Md Mohaiminul Islam, Gedas Bertasius, Mohit Bansal
https://arxiv.org/abs/2507.06485

Video-RTS: Rethinking Reinforcement Learning and Test-Time Scaling for Efficient and Enhanced Video Reasoning
Despite advances in reinforcement learning (RL)-based video reasoning with large language models (LLMs), data collection and finetuning remain significant challenges. These methods often rely on large-scale supervised fine-tuning (SFT) with extensive video data and long Chain-of-Thought (CoT) annotations, making them costly and hard to scale. To address this, we present Video-RTS, a new approach to improve video reasoning capability with drastically improved data efficiency by combining data-ef…

DrugMCTS: a drug repurposing framework combining multi-agent, RAG and Monte Carlo Tree Search
Zerui Yang, Yuwei Wan, Yinqiao Li, Yudai Matsuda, Tong Xie, Linqi Song
https://arxiv.org/abs/2507.07426

DrugMCTS: a drug repurposing framework combining multi-agent, RAG and Monte Carlo Tree Search
Recent advances in large language models have demonstrated considerable potential in scientific domains such as drug discovery. However, their effectiveness remains constrained when reasoning extends beyond the knowledge acquired during pretraining. Conventional approaches, such as fine-tuning or retrieval-augmented generation, face limitations in either imposing high computational overhead or failing to fully exploit structured scientific data. To overcome these challenges, we propose DrugMCTS…

Noise Consistency Regularization for Improved Subject-Driven Image Synthesis
Yao Ni, Song Wen, Piotr Koniusz, Anoop Cherian
https://arxiv.org/abs/2506.06483

Noise Consistency Regularization for Improved Subject-Driven Image Synthesis
Fine-tuning Stable Diffusion enables subject-driven image synthesis by adapting the model to generate images containing specific subjects. However, existing fine-tuning methods suffer from two key issues: underfitting, where the model fails to reliably capture subject identity, and overfitting, where it memorizes the subject image and reduces background diversity. To address these challenges, we propose two auxiliary consistency losses for diffusion fine-tuning. First, a prior consistency regul…

This https://arxiv.org/abs/2408.09568 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csSE_…

MergeRepair: An Exploratory Study on Merging Task-Specific Adapters in Code LLMs for Automated Program Repair
Large Language Models (LLMs) have shown high capabilities in several software development-related tasks such as program repair, documentation, code refactoring, debugging, and testing. However, training these models requires massive amount of data and significant computational resources. Adapters are specialized, small modules designed for parameter efficient fine-tuning of LLMs for specific tasks, domains, or applications without requiring extensive retraining of the entire model. These adapte…

Label-Efficient Chest X-ray Diagnosis via Partial CLIP Adaptation
Heet Nitinkumar Dalsania
https://arxiv.org/abs/2507.07254 https://a…

Label-Efficient Chest X-ray Diagnosis via Partial CLIP Adaptation
Modern deep learning implementations for medical imaging usually rely on large labeled datasets. These datasets are often difficult to obtain due to privacy concerns, high costs, and even scarcity of cases. In this paper, a label-efficient strategy is proposed for chest X-ray diagnosis that seeks to reflect real-world hospital scenarios. The experiments use the NIH Chest X-ray14 dataset and a pre-trained CLIP ViT-B/32 model. The model is adapted via partial fine-tuning of its visual encoder and…

This https://arxiv.org/abs/2506.06105 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Text-to-LoRA: Instant Transformer Adaption
While Foundation Models provide a general tool for rapid content creation, they regularly require task-specific adaptation. Traditionally, this exercise involves careful curation of datasets and repeated fine-tuning of the underlying model. Fine-tuning techniques enable practitioners to adapt foundation models for many new applications but require expensive and lengthy training while being notably sensitive to hyper-parameter choices. To overcome these limitations, we introduce Text-to-LoRA (T2…

Low-Resource Domain Adaptation for Speech LLMs via Text-Only Fine-Tuning
Yangui Fang, Jing Peng, Xu Li, Yu Xi, Chengwei Zhang, Guohui Zhong, Kai Yu
https://arxiv.org/abs/2506.05671

Low-Resource Domain Adaptation for Speech LLMs via Text-Only Fine-Tuning
Recent advances in automatic speech recognition (ASR) have combined speech encoders with large language models (LLMs) through projection, forming Speech LLMs with strong performance. However, adapting them to new domains remains challenging, especially in low-resource settings where paired speech-text data is scarce. We propose a text-only fine-tuning strategy for Speech LLMs using unpaired target-domain text without requiring additional audio. To preserve speech-text alignment, we introduce a …

Efficient Fireworks Algorithm Equipped with an Explosion Mechanism based on Student's T-distribution
Cen Shipeng, Tan Ying
https://arxiv.org/abs/2506.08484

Efficient Fireworks Algorithm Equipped with an Explosion Mechanism based on Student's T-distribution
Many real-world problems can be transformed into optimization problems, which can be classified into convex and non-convex. Although convex problems are almost completely studied in theory, many related algorithms to many non-convex problems do not work well and we need more optimization techniques. As a swarm intelligence optimization algorithm, the Fireworks Algorithm(FWA) has been widely studied and applied to many real-world scenarios, even including large language model fine-tuning. But th…

This https://arxiv.org/abs/2504.04178 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csIR_…

MSL: Not All Tokens Are What You Need for Tuning LLM as a Recommender
Large language models (LLMs), known for their comprehension capabilities and extensive knowledge, have been increasingly applied to recommendation systems (RS). Given the fundamental gap between the mechanism of LLMs and the requirement of RS, researchers have focused on fine-tuning LLMs with recommendation-specific data to enhance their performance. Language Modeling Loss (LML), originally designed for language generation tasks, is commonly adopted. However, we identify two critical limitation…

Integrating Pathology Foundation Models and Spatial Transcriptomics for Cellular Decomposition from Histology Images
Yutong Sun, Sichen Zhu, Peng Qiu
https://arxiv.org/abs/2507.07013

Integrating Pathology Foundation Models and Spatial Transcriptomics for Cellular Decomposition from Histology Images
The rapid development of digital pathology and modern deep learning has facilitated the emergence of pathology foundation models that are expected to solve general pathology problems under various disease conditions in one unified model, with or without fine-tuning. In parallel, spatial transcriptomics has emerged as a transformative technology that enables the profiling of gene expression on hematoxylin and eosin (H&E) stained histology images. Spatial transcriptomics unlocks the unprecedented…

Impact of First-order Electroweak Phase Transition on QCD Axion
Dipendu Bhandari, Soumen Kumar Manna, Arunansu Sil
https://arxiv.org/abs/2507.05353 https:/…

Impact of First-order Electroweak Phase Transition on QCD Axion
The QCD axion addresses the strong CP problem and dark matter via the misalignment mechanism, typically requiring a decay constant $f_a\sim \mathcal{O}(10^{12}$ GeV), unless the initial misalignment angle ($θ_i$) is fine-tuned. This work presents a novel approach where the possibility that the QCD axion satisfying the correct relic is extended over a broad range for $f_a\in [10^8, 10^{14}$] GeV without fine-tuning the $θ_i$, by introducing a new phase of axion oscillation dynamics across the …

Learning the Topic, Not the Language: How LLMs Classify Online Immigration Discourse Across Languages
Andrea Nasuto, Stefano Maria Iacus, Francisco Rowe, Devika Jain
https://arxiv.org/abs/2508.06435

Learning the Topic, Not the Language: How LLMs Classify Online Immigration Discourse Across Languages
Large language models (LLMs) are transforming social-science research by enabling scalable, precise analysis. Their adaptability raises the question of whether knowledge acquired through fine-tuning in a few languages can transfer to unseen languages that only appeared during pre-training. To examine this, we fine-tune lightweight LLaMA 3.2-3B models on monolingual, bilingual, or multilingual data sets to classify immigration-related tweets from X/Twitter across 13 languages, a domain character…

MagicGUI: A Foundational Mobile GUI Agent with Scalable Data Pipeline and Reinforcement Fine-tuning
Liujian Tang, Shaokang Dong, Yijia Huang, Minqi Xiang, Hongtao Ruan, Bin Wang, Shuo Li, Zhihui Cao, Hailiang Pang, Heng Kong, He Yang, Mingxu Chai, Zhilin Gao, Xingyu Liu, Yingnan Fu, Jiaming Liu, Tao Gui, Xuanjing Huang, Yu-Gang Jiang, Qi Zhang, Kang Wang, Yunke Zhang, Yuran Wang

MagicGUI: A Foundational Mobile GUI Agent with Scalable Data Pipeline and Reinforcement Fine-tuning
This paper presents MagicGUI, a foundational mobile GUI agent designed to address critical challenges in perception, grounding, and reasoning within real-world mobile GUI environments. The framework is underpinned by following six key components: (1) a comprehensive and accurate dataset, constructed via the scalable GUI Data Pipeline, which aggregates the largest and most diverse GUI-centric multimodal data to date from open-source repositories, automated crawling, and targeted manual annotatio…

TuneShield: Mitigating Toxicity in Conversational AI while Fine-tuning on Untrusted Data
Aravind Cheruvu, Shravya Kanchi, Sifat Muhammad Abdullah, Nicholas Kong, Daphne Yao, Murtuza Jadliwala, Bimal Viswanath
https://arxiv.org/abs/2507.05660

TuneShield: Mitigating Toxicity in Conversational AI while Fine-tuning on Untrusted Data
Recent advances in foundation models, such as LLMs, have revolutionized conversational AI. Chatbots are increasingly being developed by customizing LLMs on specific conversational datasets. However, mitigating toxicity during this customization, especially when dealing with untrusted training data, remains a significant challenge. To address this, we introduce TuneShield, a defense framework designed to mitigate toxicity during chatbot fine-tuning while preserving conversational quality. TuneSh…

Day 7
✅ 24 test suites, 153 tests passing.
Solid coverage across service and controller layers in my modular monorepo. Strict typing (TypeScript), full DTO validation, and realistic mocks across complex relations (TypeORM).
Next: fine-tuning error handling & exploring e2e strategies.
https://write.tyolabs.com/?p=25

Replaced article(s) found for cs.DC. https://arxiv.org/list/cs.DC/new
[1/1]:
- Fine-tuning Multimodal Transformers on Edge: A Parallel Split Learning Approach
Timo Fudala, Vasileios Tsouvalas, Nirvana Meratnia

CO-RFT: Efficient Fine-Tuning of Vision-Language-Action Models through Chunked Offline Reinforcement Learning
Dongchi Huang, Zhirui Fang, Tianle Zhang, Yihang Li, Lin Zhao, Chunhe Xia
https://arxiv.org/abs/2508.02219

CO-RFT: Efficient Fine-Tuning of Vision-Language-Action Models through Chunked Offline Reinforcement Learning
Vision-Language-Action (VLA) models demonstrate significant potential for developing generalized policies in real-world robotic control. This progress inspires researchers to explore fine-tuning these models with Reinforcement Learning (RL). However, fine-tuning VLA models with RL still faces challenges related to sample efficiency, compatibility with action chunking, and training stability. To address these challenges, we explore the fine-tuning of VLA models through offline reinforcement lear…

On the Inherent Privacy of Zeroth Order Projected Gradient Descent
Devansh Gupta, Meisam Razaviyayn, Vatsal Sharan
https://arxiv.org/abs/2507.05610 https:/…

On the Inherent Privacy of Zeroth Order Projected Gradient Descent
Differentially private zeroth-order optimization methods have recently gained popularity in private fine tuning of machine learning models due to their reduced memory requirements. Current approaches for privatizing zeroth-order methods rely on adding Gaussian noise to the estimated zeroth-order gradients. However, since the search direction in the zeroth-order methods is inherently random, researchers including Tang et al. (2024) and Zhang et al. (2024a) have raised an important question: is t…

Position: Simulating Society Requires Simulating Thought
Chance Jiajie Li, Jiayi Wu, Zhenze Mo, Ao Qu, Yuhan Tang, Kaiya Ivy Zhao, Yulu Gan, Jie Fan, Jiangbo Yu, Jinhua Zhao, Paul Liang, Luis Alonso, Kent Larson
https://arxiv.org/abs/2506.06958

Position: Simulating Society Requires Simulating Thought
Simulating society with large language models (LLMs), we argue, requires more than generating plausible behavior -- it demands cognitively grounded reasoning that is structured, revisable, and traceable. LLM-based agents are increasingly used to emulate individual and group behavior -- primarily through prompting and supervised fine-tuning. Yet they often lack internal coherence, causal reasoning, and belief traceability -- making them unreliable for analyzing how people reason, deliberate, or …

Movable Antenna Enhanced Federated Fine-Tuning of Large Language Models via Hybrid Client Selection Optimization
Yang Zhao, Yue Xiu, Chengxiao Dai, Ning Wei, Dusit Niyato
https://arxiv.org/abs/2506.00011

Movable Antenna Enhanced Federated Fine-Tuning of Large Language Models via Hybrid Client Selection Optimization
Large language model (LLM) training in 6G networks faces stringent latency and energy constraints while operating over bandwidth-limited wireless links. A commonly adopted workflow separates training into a centralized pre-training phase and a federated fine-tuning phase on domain-specific data; however, over-the-air (OTA) gradient aggregation during fine-tuning remains vulnerable to fading and interference. This study explores the integration of movable antennas (MAs), whose element positions …

CLEP-DG: Contrastive Learning for Speech Emotion Domain Generalization via Soft Prompt Tuning
Jiacheng Shi, Yanfu Zhang, Ye Gao
https://arxiv.org/abs/2507.04048

CLEP-DG: Contrastive Learning for Speech Emotion Domain Generalization via Soft Prompt Tuning
Speech Emotion Recognition (SER) is fundamental to affective computing and human-computer interaction, yet existing models struggle to generalize across diverse acoustic conditions. While Contrastive Language-Audio Pretraining (CLAP) provides strong multimodal alignment, it lacks dedicated mechanisms for capturing emotional cues, making it suboptimal for SER. To address this, we propose CLEP-DG, a framework that enhances CLAP's robustness in emotion recognition. First, we fine-tune CLAP to obta…

Fine-tuning physics-informed neural networks for cavity flows using coordinate transformation
Ryuta Takao, Satoshi Ii
https://arxiv.org/abs/2508.01122 https://

Fine-tuning physics-informed neural networks for cavity flows using coordinate transformation
Physics-informed neural networks (PINNs) have attracted attention as an alternative approach to solve partial differential equations using a deep neural network (DNN). Their simplicity and capability allow them to solve inverse problems for many applications. Despite the versatility of PINNs, it remains challenging to reduce their training cost. Using a DNN pre-trained with an arbitrary dataset with transfer learning or fine-tuning is a potential solution. However, a pre-trained model using a d…

Online Fine-Tuning of Carbon Emission Predictions using Real-Time Recurrent Learning for State Space Models
Julian Lemmel, Manuel Kranzl, Adam Lamine, Philipp Neubauer, Radu Grosu, Sophie Neubauer
https://arxiv.org/abs/2508.00804

Online Fine-Tuning of Carbon Emission Predictions using Real-Time Recurrent Learning for State Space Models
This paper introduces a new approach for fine-tuning the predictions of structured state space models (SSMs) at inference time using real-time recurrent learning. While SSMs are known for their efficiency and long-range modeling capabilities, they are typically trained offline and remain static during deployment. Our method enables online adaptation by continuously updating model parameters in response to incoming data. We evaluate our approach for linear-recurrent-unit SSMs using a small carbo…

Why LLM Safety Guardrails Collapse After Fine-tuning: A Similarity Analysis Between Alignment and Fine-tuning Datasets
Lei Hsiung, Tianyu Pang, Yung-Chen Tang, Linyue Song, Tsung-Yi Ho, Pin-Yu Chen, Yaoqing Yang
https://arxiv.org/abs/2506.05346

Why LLM Safety Guardrails Collapse After Fine-tuning: A Similarity Analysis Between Alignment and Fine-tuning Datasets
Recent advancements in large language models (LLMs) have underscored their vulnerability to safety alignment jailbreaks, particularly when subjected to downstream fine-tuning. However, existing mitigation strategies primarily focus on reactively addressing jailbreak incidents after safety guardrails have been compromised, removing harmful gradients during fine-tuning, or continuously reinforcing safety alignment throughout fine-tuning. As such, they tend to overlook a critical upstream factor: …

This https://arxiv.org/abs/2506.05673 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Peer-Ranked Precision: Creating a Foundational Dataset for Fine-Tuning Vision Models from DataSeeds' Annotated Imagery
The development of modern Artificial Intelligence (AI) models, particularly diffusion-based models employed in computer vision and image generation tasks, is undergoing a paradigmatic shift in development methodologies. Traditionally dominated by a "Model Centric" approach, in which performance gains were primarily pursued through increasingly complex model architectures and hyperparameter optimization, the field is now recognizing a more nuanced "Data-Centric" approach. This emergent framework…

Gradient Inversion Attacks on Parameter-Efficient Fine-Tuning
Hasin Us Sami, Swapneel Sen, Amit K. Roy-Chowdhury, Srikanth V. Krishnamurthy, Basak Guler
https://arxiv.org/abs/2506.04453

Gradient Inversion Attacks on Parameter-Efficient Fine-Tuning
Federated learning (FL) allows multiple data-owners to collaboratively train machine learning models by exchanging local gradients, while keeping their private data on-device. To simultaneously enhance privacy and training efficiency, recently parameter-efficient fine-tuning (PEFT) of large-scale pretrained models has gained substantial attention in FL. While keeping a pretrained (backbone) model frozen, each user fine-tunes only a few lightweight modules to be used in conjunction, to fit speci…

A foundation model to predict and capture human cognition.
https://www.nature.com/articles/s41586-025-09215-4
And the reaction and criticism,

A foundation model to predict and capture human cognition - Nature
A computational model called Centaur, developed by fine-tuning a language model on a huge dataset called Psych-101, can predict and simulate human nature in experiments expressible in natural language, even in previously unseen situations.

Fine-Tuning LLMs For ‘Good’ Behavior Makes Them More Likely To Say No https://www.404media.co/fine-tuning-llms-cognitive-bias-decision-making-study/

GradOT: Training-free Gradient-preserving Offsite-tuning for Large Language Models
Kai Yao, Zhaorui Tan, Penglei Gao, Lichun Li, Kaixin Wu, Yinggui Wang, Yuan Zhao, Yixin Ji, Wei Wang, Jianke Zhu
https://arxiv.org/abs/2507.04455

GradOT: Training-free Gradient-preserving Offsite-tuning for Large Language Models
The rapid growth of large language models (LLMs) with traditional centralized fine-tuning emerges as a key technique for adapting these models to domain-specific challenges, yielding privacy risks for both model and data owners. One promising solution, called offsite-tuning (OT), is proposed to address these challenges, where a weaker emulator is compressed from the original model and further fine-tuned with adapter to enhance privacy. However, the existing OT-based methods require high computa…

Hear-Your-Click: Interactive Video-to-Audio Generation via Object-aware Contrastive Audio-Visual Fine-tuning
Yingshan Liang, Keyu Fan, Zhicheng Du, Yiran Wang, Qingyang Shi, Xinyu Zhang, Jiasheng Lu, Peiwu Qin
https://arxiv.org/abs/2507.04959

Hear-Your-Click: Interactive Video-to-Audio Generation via Object-aware Contrastive Audio-Visual Fine-tuning
Video-to-audio (V2A) generation shows great potential in fields such as film production. Despite significant advances, current V2A methods, which rely on global video information, struggle with complex scenes and often fail to generate audio tailored to specific objects or regions in the videos. To address these limitations, we introduce Hear-Your-Click, an interactive V2A framework that enables users to generate sounds for specific objects in the videos by simply clicking on the frame. To achi…

Understanding Overadaptation in Supervised Fine-Tuning: The Role of Ensemble Methods
Yifan Hao, Xingyuan Pan, Hanning Zhang, Chenlu Ye, Rui Pan, Tong Zhang
https://arxiv.org/abs/2506.01901

Understanding Overadaptation in Supervised Fine-Tuning: The Role of Ensemble Methods
Supervised fine-tuning (SFT) on domain-specific data is the dominant approach for adapting foundation models to specialized tasks. However, it has been observed that SFT models tend to forget knowledge acquired during pretraining. In vision models, ensembling a pretrained model with its fine-tuned counterpart has been shown to mitigate this issue. In this work, we demonstrate that the same holds for language models, and, more strikingly, we observe an overadaptation phenomenon: the ensemble mod…

This https://arxiv.org/abs/2506.02916 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csIR_…

MMM4Rec: An Transfer-Efficient Framework for Multi-modal Sequential Recommendation
Sequential Recommendation (SR) systems model user preferences by analyzing interaction histories. Although transferable multi-modal SR architectures demonstrate superior performance compared to traditional ID-based approaches, current methods incur substantial fine-tuning costs when adapting to new domains due to complex optimization requirements and negative transfer effects - a significant deployment bottleneck that hinders engineers from efficiently repurposing pre-trained models for novel a…

VisCoder: Fine-Tuning LLMs for Executable Python Visualization Code Generation
Yuansheng Ni, Ping Nie, Kai Zou, Xiang Yue, Wenhu Chen
https://arxiv.org/abs/2506.03930

VisCoder: Fine-Tuning LLMs for Executable Python Visualization Code Generation
Large language models (LLMs) often struggle with visualization tasks like plotting diagrams, charts, where success depends on both code correctness and visual semantics. Existing instruction-tuning datasets lack execution-grounded supervision and offer limited support for iterative code correction, resulting in fragile and unreliable plot generation. We present VisCode-200K, a large-scale instruction tuning dataset for Python-based visualization and self-correction. It contains over 200K exampl…

This https://arxiv.org/abs/2505.01997 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Restoring Calibration for Aligned Large Language Models: A Calibration-Aware Fine-Tuning Approach
One of the key technologies for the success of Large Language Models (LLMs) is preference alignment. However, a notable side effect of preference alignment is poor calibration: while the pre-trained models are typically well-calibrated, LLMs tend to become poorly calibrated after alignment with human preferences. In this paper, we investigate why preference alignment affects calibration and how to address this issue. For the first question, we observe that the preference collapse issue in align…

Analysis and Optimized CXL-Attached Memory Allocation for Long-Context LLM Fine-Tuning
Yong-Cheng Liaw, Shuo-Han Chen
https://arxiv.org/abs/2507.03305 http…

Analysis and Optimized CXL-Attached Memory Allocation for Long-Context LLM Fine-Tuning
The growing prevalence of Large Language Models (LLMs) and their substantial memory requirements have prompted renewed interest in CPU offloading as a method to compensate for limited GPU memory. In particular, when CPU memory is leveraged to temporarily store intermediate states of LLMs, CPU memory becomes a new bottleneck and soon reaches the capacity limitation of commodity CPUs. In this work, we investigate the effectiveness of Compute Express Link (CXL) add-in card (AIC) memory as an exten…

This https://arxiv.org/abs/2505.22942 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCL_…

WorkForceAgent-R1: Incentivizing Reasoning Capability in LLM-based Web Agents via Reinforcement Learning
Large language models (LLMs)-empowered web agents enables automating complex, real-time web navigation tasks in enterprise environments. However, existing web agents relying on supervised fine-tuning (SFT) often struggle with generalization and robustness due to insufficient reasoning capabilities when handling the inherently dynamic nature of web interactions. In this study, we introduce WorkForceAgent-R1, an LLM-based web agent trained using a rule-based R1-style reinforcement learning framew…

FedShield-LLM: A Secure and Scalable Federated Fine-Tuned Large Language Model
Md Jueal Mia, M. Hadi Amini
https://arxiv.org/abs/2506.05640 https://…

FedShield-LLM: A Secure and Scalable Federated Fine-Tuned Large Language Model
Federated Learning (FL) offers a decentralized framework for training and fine-tuning Large Language Models (LLMs) by leveraging computational resources across organizations while keeping sensitive data on local devices. It addresses privacy and security concerns while navigating challenges associated with the substantial computational demands of LLMs, which can be prohibitive for small and medium-sized organizations. FL supports the development of task-specific LLMs for cross-silo applications…

Approaching Dialogue State Tracking via Aligning Speech Encoders and LLMs
\v{S}imon Sedl\'a\v{c}ek, Bolaji Yusuf, J\'an \v{S}vec, Pradyoth Hegde, Santosh Kesiraju, Old\v{r}ich Plchot, Jan \v{C}ernock\'y
https://arxiv.org/abs/2506.08633

Approaching Dialogue State Tracking via Aligning Speech Encoders and LLMs
In this work, we approach spoken Dialogue State Tracking (DST) by bridging the representation spaces of speech encoders and LLMs via a small connector module, with a focus on fully open-sourced and open-data components (WavLM-large, OLMo). We focus on ablating different aspects of such systems including full/LoRA adapter fine-tuning, the effect of agent turns in the dialogue history, as well as fuzzy matching-based output post-processing, which greatly improves performance of our systems on nam…

This https://arxiv.org/abs/2505.23868 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Noise-Robustness Through Noise: Asymmetric LoRA Adaption with Poisoning Expert
Current parameter-efficient fine-tuning methods for adapting pre-trained language models to downstream tasks are susceptible to interference from noisy data. Conventional noise-handling approaches either rely on laborious data pre-processing or employ model architecture modifications prone to error accumulation. In contrast to existing noise-process paradigms, we propose a noise-robust adaptation method via asymmetric LoRA poisoning experts (LoPE), a novel framework that enhances model robustne…

TTS-CtrlNet: Time varying emotion aligned text-to-speech generation with ControlNet
Jaeseok Jeong, Yuna Lee, Mingi Kwon, Youngjung Uh
https://arxiv.org/abs/2507.04349

TTS-CtrlNet: Time varying emotion aligned text-to-speech generation with ControlNet
Recent advances in text-to-speech (TTS) have enabled natural speech synthesis, but fine-grained, time-varying emotion control remains challenging. Existing methods often allow only utterance-level control and require full model fine-tuning with a large emotion speech dataset, which can degrade performance. Inspired by adding conditional control to the existing model in ControlNet (Zhang et al, 2023), we propose the first ControlNet-based approach for controllable flow-matching TTS (TTS-CtrlNet)…

Post-training for Efficient Communication via Convention Formation
Yilun Hua, Evan Wang, Yoav Artzi
https://arxiv.org/abs/2508.06482 https://arxiv.org/pdf/…

Post-training for Efficient Communication via Convention Formation
Humans communicate with increasing efficiency in multi-turn interactions, by adapting their language and forming ad-hoc conventions. In contrast, prior work shows that LLMs do not naturally show this behavior. We develop a post-training process to develop this ability through targeted fine-tuning on heuristically identified demonstrations of convention formation. We evaluate with two new benchmarks focused on this capability. First, we design a focused, cognitively-motivated interaction benchma…

This https://arxiv.org/abs/2505.00347 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

Pushing the Limits of Low-Bit Optimizers: A Focus on EMA Dynamics
The rapid scaling of models has led to prohibitively high training and fine-tuning costs. A major factor accounting for memory consumption is the widespread use of stateful optimizers (e.g., Adam), which maintain auxiliary information of even 2x the model size in order to achieve optimal convergence. We therefore present SOLO in this work to spawn a novel type of optimizer that requires an extremely light memory footprint. While previous efforts have achieved certain success in 8-bit or 4-bit c…

Your Agent Can Defend Itself against Backdoor Attacks
Li Changjiang, Liang Jiacheng, Cao Bochuan, Chen Jinghui, Wang Ting
https://arxiv.org/abs/2506.08336 …

Your Agent Can Defend Itself against Backdoor Attacks
Despite their growing adoption across domains, large language model (LLM)-powered agents face significant security risks from backdoor attacks during training and fine-tuning. These compromised agents can subsequently be manipulated to execute malicious operations when presented with specific triggers in their inputs or environments. To address this pressing risk, we present ReAgent, a novel defense against a range of backdoor attacks on LLM-based agents. Intuitively, backdoor attacks often res…

Multi-Agent Debate Strategies to Enhance Requirements Engineering with Large Language Models
Marc Oriol, Quim Motger, Jordi Marco, Xavier Franch
https://arxiv.org/abs/2507.05981

Multi-Agent Debate Strategies to Enhance Requirements Engineering with Large Language Models
Context: Large Language Model (LLM) agents are becoming widely used for various Requirements Engineering (RE) tasks. Research on improving their accuracy mainly focuses on prompt engineering, model fine-tuning, and retrieval augmented generation. However, these methods often treat models as isolated black boxes - relying on single-pass outputs without iterative refinement or collaboration, limiting robustness and adaptability. Objective: We propose that, just as human debates enhance accuracy a…

Are You Listening to Me? Fine-Tuning Chatbots for Empathetic Dialogue
Paulo Ricardo Knob, Leonardo Scholler, Juliano Rigatti, Soraia Raupp Musse
https://arxiv.org/abs/2507.02537

Are You Listening to Me? Fine-Tuning Chatbots for Empathetic Dialogue
Conversational agents have made significant progress since ELIZA, expanding their role across various domains, including healthcare, education, and customer service. As these agents become increasingly integrated into daily human interactions, the need for emotional intelligence, particularly empathetic listening, becomes increasingly essential. In this study, we explore how Large Language Models (LLMs) respond when tasked with generating emotionally rich interactions. Starting from a small dat…

Enhancing Food-Domain Question Answering with a Multimodal Knowledge Graph: Hybrid QA Generation and Diversity Analysis
Srihari K B, Pushpak Bhattacharyya
https://arxiv.org/abs/2507.06571

Enhancing Food-Domain Question Answering with a Multimodal Knowledge Graph: Hybrid QA Generation and Diversity Analysis
We propose a unified food-domain QA framework that combines a large-scale multimodal knowledge graph (MMKG) with generative AI. Our MMKG links 13,000 recipes, 3,000 ingredients, 140,000 relations, and 14,000 images. We generate 40,000 QA pairs using 40 templates and LLaVA/DeepSeek augmentation. Joint fine-tuning of Meta LLaMA 3.1-8B and Stable Diffusion 3.5-Large improves BERTScore by 16.2\%, reduces FID by 37.8\%, and boosts CLIP alignment by 31.1\%. Diagnostic analyses-CLIP-based mismatch det…

This https://arxiv.org/abs/2506.01790 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

$IF-GUIDE$: Influence Function-Guided Detoxification of LLMs
We study how training data contributes to the emergence of toxic behaviors in large-language models. Most prior work on reducing model toxicity adopts $reactive$ approaches, such as fine-tuning pre-trained (and potentially toxic) models to align them with human values. In contrast, we propose a $proactive$ approach$-$IF-Guide$-$which leverages influence functions to identify harmful tokens within any training data and suppress their impact during training. To this end, we first show that standa…

DiWA: Diffusion Policy Adaptation with World Models
Akshay L Chandra, Iman Nematollahi, Chenguang Huang, Tim Welschehold, Wolfram Burgard, Abhinav Valada
https://arxiv.org/abs/2508.03645

DiWA: Diffusion Policy Adaptation with World Models
Fine-tuning diffusion policies with reinforcement learning (RL) presents significant challenges. The long denoising sequence for each action prediction impedes effective reward propagation. Moreover, standard RL methods require millions of real-world interactions, posing a major bottleneck for practical fine-tuning. Although prior work frames the denoising process in diffusion policies as a Markov Decision Process to enable RL-based updates, its strong dependence on environment interaction rema…

This https://arxiv.org/abs/2506.05394 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCR_…

Attacking Attention of Foundation Models Disrupts Downstream Tasks
Foundation models represent the most prominent and recent paradigm shift in artificial intelligence.Foundation models are large models, trained on broad data that deliver high accuracy in many downstream tasks, often without fine-tuning. For this reason, models such as CLIP , DINO or Vision Transfomers (ViT), are becoming the bedrock of many industrial AI-powered applications. However, the reliance on pre-trained foundation models also introduces significant security concerns, as these models a…

This https://arxiv.org/abs/2410.00527 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_ees…

Wanna hear your voice? A sample is all we need!
Research on audio clue-based target speaker extraction (TSE) has focused on modeling mixtures and reference speech, achieving strong results in English due to abundant datasets. However, cross-lingual properties remain underexplored, as low-resource languages face challenges from limited annotated data and linguistic resources. To bridge this gap, we propose WHYV (Wanna Hear Your Voice), a cross-lingual TSE framework enabling zero-shot adaptation without fine-tuning. WHYV employs a frequency-mo…

Memory-Efficient Split Federated Learning for LLM Fine-Tuning on Heterogeneous Mobile Devices
Xiaopei Chen, Liang Li, Fei Ji, Wen Wu
https://arxiv.org/abs/2506.02940

Memory-Efficient Split Federated Learning for LLM Fine-Tuning on Heterogeneous Mobile Devices
In this paper, we propose an edge-assisted split federated learning framework to facilitate large language model (LLM) fine-tuning on heterogeneous mobile devices while alleviating memory pressures on both mobile devices and the edge server. Specifically, mobile devices perform low-rank adaptation (LoRA) fine-tuning on only a subset of lower layers of the pre-trained LLM, tailored to their individual capacities. On the server, a full LLM is maintained, and the corresponding LoRA modules are sel…

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[4/5]:
- Fine-tuning Multimodal Transformers on Edge: A Parallel Split Learning Approach
Timo Fudala, Vasileios Tsouvalas, Nirvana Meratnia

Checklist Engineering Empowers Multilingual LLM Judges
Mohammad Ghiasvand Mohammadkhani, Hamid Beigy
https://arxiv.org/abs/2507.06774 https://

Checklist Engineering Empowers Multilingual LLM Judges
Automated text evaluation has long been a central issue in Natural Language Processing (NLP). Recently, the field has shifted toward using Large Language Models (LLMs) as evaluators-a trend known as the LLM-as-a-Judge paradigm. While promising and easily adaptable across tasks, this approach has seen limited exploration in multilingual contexts. Existing multilingual studies often rely on proprietary models or require extensive training data for fine-tuning, raising concerns about cost, time, a…

This https://arxiv.org/abs/2505.21920 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCV_…

InfoSAM: Fine-Tuning the Segment Anything Model from An Information-Theoretic Perspective
The Segment Anything Model (SAM), a vision foundation model, exhibits impressive zero-shot capabilities in general tasks but struggles in specialized domains. Parameter-efficient fine-tuning (PEFT) is a promising approach to unleash the potential of SAM in novel scenarios. However, existing PEFT methods for SAM neglect the domain-invariant relations encoded in the pre-trained model. To bridge this gap, we propose InfoSAM, an information-theoretic approach that enhances SAM fine-tuning by distil…

Fine-Tuning Text-to-Speech Diffusion Models Using Reinforcement Learning with Human Feedback
Jingyi Chen, Ju Seung Byun, Micha Elsner, Pichao Wang, Andrew Perrault
https://arxiv.org/abs/2508.03123

Fine-Tuning Text-to-Speech Diffusion Models Using Reinforcement Learning with Human Feedback
Diffusion models produce high-fidelity speech but are inefficient for real-time use due to long denoising steps and challenges in modeling intonation and rhythm. To improve this, we propose Diffusion Loss-Guided Policy Optimization (DLPO), an RLHF framework for TTS diffusion models. DLPO integrates the original training loss into the reward function, preserving generative capabilities while reducing inefficiencies. Using naturalness scores as feedback, DLPO aligns reward optimization with the d…

SafeCOMM: What about Safety Alignment in Fine-Tuned Telecom Large Language Models?
Aladin Djuhera, Swanand Ravindra Kadhe, Farhan Ahmed, Syed Zawad, Holger Boche, Walid Saad
https://arxiv.org/abs/2506.00062

SafeCOMM: What about Safety Alignment in Fine-Tuned Telecom Large Language Models?
Fine-tuning large language models (LLMs) for telecom tasks and datasets is a common practice to adapt general-purpose models to the telecom domain. However, little attention has been paid to how this process may compromise model safety. Recent research has shown that even benign fine-tuning can degrade the safety alignment of LLMs, causing them to respond to harmful or unethical user queries. In this paper, we investigate this issue for telecom-tuned LLMs using three representative datasets fea…

Text-to-LoRA: Instant Transformer Adaption
Rujikorn Charakorn, Edoardo Cetin, Yujin Tang, Robert Tjarko Lange
https://arxiv.org/abs/2506.06105 https://

Text-to-LoRA: Instant Transformer Adaption
While Foundation Models provide a general tool for rapid content creation, they regularly require task-specific adaptation. Traditionally, this exercise involves careful curation of datasets and repeated fine-tuning of the underlying model. Fine-tuning techniques enable practitioners to adapt foundation models for many new applications but require expensive and lengthy training while being notably sensitive to hyper-parameter choices. To overcome these limitations, we introduce Text-to-LoRA (T2…

WGLE:Backdoor-free and Multi-bit Black-box Watermarking for Graph Neural Networks
Tingzhi Li, Xuefeng Liu
https://arxiv.org/abs/2506.08602 https://

WGLE:Backdoor-free and Multi-bit Black-box Watermarking for Graph Neural Networks
Graph Neural Networks (GNNs) are increasingly deployed in graph-related applications, making ownership verification critical to protect their intellectual property against model theft. Fingerprinting and black-box watermarking are two main methods. However, the former relies on determining model similarity, which is computationally expensive and prone to ownership collisions after model post-processing such as model pruning or fine-tuning. The latter embeds backdoors, exposing watermarked model…

Whispers of Many Shores: Cultural Alignment through Collaborative Cultural Expertise
Shuai Feng, Wei-Chuang Chan, Srishti Chouhan, Junior Francisco Garcia Ayala, Srujananjali Medicherla, Kyle Clark, Mingwei Shi
https://arxiv.org/abs/2506.00242

Whispers of Many Shores: Cultural Alignment through Collaborative Cultural Expertise
The integration of large language models (LLMs) into global applications necessitates effective cultural alignment for meaningful and culturally-sensitive interactions. Current LLMs often lack the nuanced understanding required for diverse cultural contexts, and adapting them typically involves costly full fine-tuning. To address this, we introduce a novel soft prompt fine-tuning framework that enables efficient and modular cultural alignment. Our method utilizes vectorized prompt tuning to dyn…

EXPO: Stable Reinforcement Learning with Expressive Policies
Perry Dong, Qiyang Li, Dorsa Sadigh, Chelsea Finn
https://arxiv.org/abs/2507.07986 https://arxiv.org/pdf/2507.07986 https://arxiv.org/html/2507.07986
arXiv:2507.07986v1 Announce Type: new
Abstract: We study the problem of training and fine-tuning expressive policies with online reinforcement learning (RL) given an offline dataset. Training expressive policy classes with online RL present a unique challenge of stable value maximization. Unlike simpler Gaussian policies commonly used in online RL, expressive policies like diffusion and flow-matching policies are parameterized by a long denoising chain, which hinders stable gradient propagation from actions to policy parameters when optimizing against some value function. Our key insight is that we can address stable value maximization by avoiding direct optimization over value with the expressive policy and instead construct an on-the-fly RL policy to maximize Q-value. We propose Expressive Policy Optimization (EXPO), a sample-efficient online RL algorithm that utilizes an on-the-fly policy to maximize value with two parameterized policies -- a larger expressive base policy trained with a stable imitation learning objective and a light-weight Gaussian edit policy that edits the actions sampled from the base policy toward a higher value distribution. The on-the-fly policy optimizes the actions from the base policy with the learned edit policy and chooses the value maximizing action from the base and edited actions for both sampling and temporal-difference (TD) backup. Our approach yields up to 2-3x improvement in sample efficiency on average over prior methods both in the setting of fine-tuning a pretrained policy given offline data and in leveraging offline data to train online.
toXiv_bot_toot

RecRankerEval: A Flexible and Extensible Framework for Top-k LLM-based Recommendation
Zeyuan Meng, Zixuan Yi, Iadh Ounis
https://arxiv.org/abs/2507.05880 h…

RecRankerEval: A Flexible and Extensible Framework for Top-k LLM-based Recommendation
A recent Large language model (LLM)-based recommendation model, called RecRanker, has demonstrated a superior performance in the top-k recommendation task compared to other models. In particular, RecRanker samples users via clustering, generates an initial ranking list using an initial recommendation model, and fine-tunes an LLM through hybrid instruction tuning to infer user preferences. However, the contribution of each core component remains underexplored. In this work, we inspect the reprod…

EcoLoRA: Communication-Efficient Federated Fine-Tuning of Large Language Models
Han Liu, Ruoyao Wen, Srijith Nair, Jia Liu, Wenjing Lou, Chongjie Zhang, William Yeoh, Yevgeniy Vorobeychik, Ning Zhang
https://arxiv.org/abs/2506.02001

EcoLoRA: Communication-Efficient Federated Fine-Tuning of Large Language Models
To address data locality and privacy restrictions, Federated Learning (FL) has recently been adopted to fine-tune large language models (LLMs), enabling improved performance on various downstream tasks without requiring aggregated data. However, the repeated exchange of model updates in FL can result in prohibitively high communication costs, hindering the distributed learning process. To address this challenge, we propose EcoLoRA, a novel communication-efficient federated fine-tuning framework…

EXPOTION: Facial Expression and Motion Control for Multimodal Music Generation
Fathinah Izzati, Xinyue Li, Gus Xia
https://arxiv.org/abs/2507.04955 https:/…

EXPOTION: Facial Expression and Motion Control for Multimodal Music Generation
We propose Expotion (Facial Expression and Motion Control for Multimodal Music Generation), a generative model leveraging multimodal visual controls - specifically, human facial expressions and upper-body motion - as well as text prompts to produce expressive and temporally accurate music. We adopt parameter-efficient fine-tuning (PEFT) on the pretrained text-to-music generation model, enabling fine-grained adaptation to the multimodal controls using a small dataset. To ensure precise synchroni…

Training-free Generation of Temporally Consistent Rewards from VLMs
Yinuo Zhao, Jiale Yuan, Zhiyuan Xu, Xiaoshuai Hao, Xinyi Zhang, Kun Wu, Zhengping Che, Chi Harold Liu, Jian Tang
https://arxiv.org/abs/2507.04789

Training-free Generation of Temporally Consistent Rewards from VLMs
Recent advances in vision-language models (VLMs) have significantly improved performance in embodied tasks such as goal decomposition and visual comprehension. However, providing accurate rewards for robotic manipulation without fine-tuning VLMs remains challenging due to the absence of domain-specific robotic knowledge in pre-trained datasets and high computational costs that hinder real-time applicability. To address this, we propose $\mathrm{T}^2$-VLM, a novel training-free, temporally consi…

AdS: Adapter-state Sharing Framework for Multimodal Sarcasm Detection
Soumyadeep Jana, Sahil Danayak, Sanasam Ranbir Singh
https://arxiv.org/abs/2507.04508

AdS: Adapter-state Sharing Framework for Multimodal Sarcasm Detection
The growing prevalence of multimodal image-text sarcasm on social media poses challenges for opinion mining, especially under resource constraints. Existing approaches rely on full fine-tuning of large pre-trained models, making them unsuitable for low-resource settings. While recent parameter-efficient fine-tuning (PEFT) methods offer promise, their off-the-shelf use underperforms on complex tasks like sarcasm detection. We propose AdS (Adapter-State Sharing), a lightweight framework built on …

GORACS: Group-level Optimal Transport-guided Coreset Selection for LLM-based Recommender Systems
Tiehua Mei, Hengrui Chen, Peng Yu, Jiaqing Liang, Deqing Yang
https://arxiv.org/abs/2506.04015

GORACS: Group-level Optimal Transport-guided Coreset Selection for LLM-based Recommender Systems
Although large language models (LLMs) have shown great potential in recommender systems, the prohibitive computational costs for fine-tuning LLMs on entire datasets hinder their successful deployment in real-world scenarios. To develop affordable and effective LLM-based recommender systems, we focus on the task of coreset selection which identifies a small subset of fine-tuning data to optimize the test loss, thereby facilitating efficient LLMs' fine-tuning. Although there exist some intuitive …

Open Vision Reasoner: Transferring Linguistic Cognitive Behavior for Visual Reasoning
Yana Wei, Liang Zhao, Jianjian Sun, Kangheng Lin, Jisheng Yin, Jingcheng Hu, Yinmin Zhang, En Yu, Haoran Lv, Zejia Weng, Jia Wang, Chunrui Han, Yuang Peng, Qi Han, Zheng Ge, Xiangyu Zhang, Daxin Jiang, Vishal M. Patel
https://arxiv.org/abs/2507…

Open Vision Reasoner: Transferring Linguistic Cognitive Behavior for Visual Reasoning
The remarkable reasoning capability of large language models (LLMs) stems from cognitive behaviors that emerge through reinforcement with verifiable rewards. This work investigates how to transfer this principle to Multimodal LLMs (MLLMs) to unlock advanced visual reasoning. We introduce a two-stage paradigm built on Qwen2.5-VL-7B: a massive linguistic cold-start fine-tuning, followed by multimodal reinforcement learning (RL) spanning nearly 1,000 steps, surpassing all previous open-source effo…

Impact of Fine-Tuning Methods on Memorization in Large Language Models
Jie Hou, Chuxiong Wu, Lannan Luo, Qiang Zeng
https://arxiv.org/abs/2507.00258 https:…

Impact of Fine-Tuning Methods on Memorization in Large Language Models
As the capabilities of pre-trained large language models (LLMs) continue to advance, the "pre-train and fine-tune" paradigm has become increasingly mainstream, leading to the development of various fine-tuning methods. However, the privacy risks arising from memorization during fine-tuning have received relatively little attention. To address this gap, we categorize popular fine-tuning approaches and assess their impact on memorization through the lens of membership inference attacks (MIAs). Ou…

This https://arxiv.org/abs/2505.24200 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csSD_…

Improving Multilingual Speech Models on ML-SUPERB 2.0: Fine-tuning with Data Augmentation and LID-Aware CTC
Multilingual speech processing with self-supervised or supervised pre-trained Speech Foundation Models (SFM) has achieved strong performance on tasks like Language Identification (LID) and Automatic Speech Recognition (ASR). However, these models struggle with limited resources during fine-tuning. This paper enhances multilingual LID and ASR on ML-SUPERB 2.0 by exploring multiple strategies for adapting SFMs, including frozen upstream training, partial fine-tuning, and low-rank adaptation. Furt…

Enhancing Safety of Foundation Models for Visual Navigation through Collision Avoidance via Repulsive Estimation
Joonkyung Kim, Joonyeol Sim, Woojun Kim, Katia Sycara, Changjoo Nam
https://arxiv.org/abs/2506.03834

Enhancing Safety of Foundation Models for Visual Navigation through Collision Avoidance via Repulsive Estimation
We propose CARE (Collision Avoidance via Repulsive Estimation), a plug-and-play module that enhances the safety of vision-based navigation without requiring additional range sensors or fine-tuning of pretrained models. While recent foundation models using only RGB inputs have shown strong performance, they often fail to generalize in out-of-distribution (OOD) environments with unseen objects or variations in camera parameters (e.g., field of view, pose, or focal length). Without fine-tuning, th…

SU-ESRGAN: Semantic and Uncertainty-Aware ESRGAN for Super-Resolution of Satellite and Drone Imagery with Fine-Tuning for Cross Domain Evaluation
Prerana Ramkumar
https://arxiv.org/abs/2508.00750

SU-ESRGAN: Semantic and Uncertainty-Aware ESRGAN for Super-Resolution of Satellite and Drone Imagery with Fine-Tuning for Cross Domain Evaluation
Generative Adversarial Networks (GANs) have achieved realistic super-resolution (SR) of images however, they lack semantic consistency and per-pixel confidence, limiting their credibility in critical remote sensing applications such as disaster response, urban planning and agriculture. This paper introduces Semantic and Uncertainty-Aware ESRGAN (SU-ESRGAN), the first SR framework designed for satellite imagery to integrate the ESRGAN, segmentation loss via DeepLabv3 for class detail preservatio…

Fine-Tuning ASR for Stuttered Speech: Personalized vs. Generalized Approaches
Dena Mujtaba, Nihar Mahapatra
https://arxiv.org/abs/2506.00853 https://

Fine-Tuning ASR for Stuttered Speech: Personalized vs. Generalized Approaches
Stuttering -- characterized by involuntary disfluencies such as blocks, prolongations, and repetitions -- is often misinterpreted by automatic speech recognition (ASR) systems, resulting in elevated word error rates and making voice-driven technologies inaccessible to people who stutter. The variability of disfluencies across speakers and contexts further complicates ASR training, compounded by limited annotated stuttered speech data. In this paper, we investigate fine-tuning ASRs for stuttered…

MLLM-Fabric: Multimodal Large Language Model-Driven Robotic Framework for Fabric Sorting and Selection
Liman Wang, Hanyang Zhong, Tianyuan Wang, Shan Luo, Jihong Zhu
https://arxiv.org/abs/2507.04351

MLLM-Fabric: Multimodal Large Language Model-Driven Robotic Framework for Fabric Sorting and Selection
Choosing the right fabric is crucial to meet functional and quality requirements in robotic applications for textile manufacturing, apparel production, and smart retail. We present MLLM-Fabric, a robotic framework powered by multimodal large language models (MLLMs) for fabric sorting and selection. The system includes a robotic arm, a camera, a visuotactile sensor, and a pressure sensor. It employs supervised fine-tuning and multimodal explanation-guided knowledge distillation to accurately cla…

Sentinel: SOTA model to protect against prompt injections
Dror Ivry, Oran Nahum
https://arxiv.org/abs/2506.05446 https://arxiv.org/pd…

Sentinel: SOTA model to protect against prompt injections
Large Language Models (LLMs) are increasingly powerful but remain vulnerable to prompt injection attacks, where malicious inputs cause the model to deviate from its intended instructions. This paper introduces Sentinel, a novel detection model, qualifire/prompt-injection-sentinel, based on the \answerdotai/ModernBERT-large architecture. By leveraging ModernBERT's advanced features and fine-tuning on an extensive and diverse dataset comprising a few open-source and private collections, Sentinel …

Navigating Sparse Molecular Data with Stein Diffusion Guidance
Van Khoa Nguyen, Lionel Blond\'e, Alexandros Kalousis
https://arxiv.org/abs/2507.05482 h…

Navigating Sparse Molecular Data with Stein Diffusion Guidance
Stochastic optimal control (SOC) has recently emerged as a principled framework for fine-tuning diffusion models. However, its dependence on computationally intensive simulations makes it impractical for fast sampling. In parallel, a class of training-free approaches has been developed that guides diffusion models using off-the-shelf classifiers on predicted clean samples, bypassing the need to train classifiers on noisy data. These methods can be interpreted as approximate SOC schemes, using T…

TR-PTS: Task-Relevant Parameter and Token Selection for Efficient Tuning
Siqi Luo, Haoran Yang, Yi Xin, Mingyang Yi, Guangyang Wu, Guangtao Zhai, Xiaohong Liu
https://arxiv.org/abs/2507.22872

TR-PTS: Task-Relevant Parameter and Token Selection for Efficient Tuning
Large pre-trained models achieve remarkable performance in vision tasks but are impractical for fine-tuning due to high computational and storage costs. Parameter-Efficient Fine-Tuning (PEFT) methods mitigate this issue by updating only a subset of parameters; however, most existing approaches are task-agnostic, failing to fully exploit task-specific adaptations, which leads to suboptimal efficiency and performance. To address this limitation, we propose Task-Relevant Parameter and Token Select…

Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learning
Yuheng Lei, Sitong Mao, Shunbo Zhou, Hongyuan Zhang, Xuelong Li, Ping Luo
https://arxiv.org/abs/2506.05985

Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learning
A generalist agent must continuously learn and adapt throughout its lifetime, achieving efficient forward transfer while minimizing catastrophic forgetting. Previous work within the dominant pretrain-then-finetune paradigm has explored parameter-efficient fine-tuning for single-task adaptation, effectively steering a frozen pretrained model with a small number of parameters. However, in the context of lifelong learning, these methods rely on the impractical assumption of a test-time task identi…

Learning to Diagnose Privately: DP-Powered LLMs for Radiology Report Classification
Payel Bhattacharjee, Fengwei Tian, Ravi Tandon, Joseph Lo, Heidi Hanson, Geoffrey Rubin, Nirav Merchant, John Gounley
https://arxiv.org/abs/2506.04450

Learning to Diagnose Privately: DP-Powered LLMs for Radiology Report Classification
Purpose: This study proposes a framework for fine-tuning large language models (LLMs) with differential privacy (DP) to perform multi-abnormality classification on radiology report text. By injecting calibrated noise during fine-tuning, the framework seeks to mitigate the privacy risks associated with sensitive patient data and protect against data leakage while maintaining classification performance. Materials and Methods: We used 50,232 radiology reports from the publicly available MIMIC-CXR …

Corrector Sampling in Language Models
Itai Gat, Neta Shaul, Uriel Singer, Yaron Lipman
https://arxiv.org/abs/2506.06215 https://arxiv…

Corrector Sampling in Language Models
Autoregressive language models accumulate errors due to their fixed, irrevocable left-to-right token generation. To address this, we propose a new sampling method called Resample-Previous-Tokens (RPT). RPT mitigates error accumulation by iteratively revisiting and potentially replacing tokens in a window of previously generated text. This method can be integrated into existing autoregressive models, preserving their next-token-prediction quality and speed. Fine-tuning a pretrained 8B parameter …

Fine-Tuning Lowers Safety and Disrupts Evaluation Consistency
Kathleen C. Fraser, Hillary Dawkins, Isar Nejadgholi, Svetlana Kiritchenko
https://arxiv.org/abs/2506.17209

Fine-Tuning Lowers Safety and Disrupts Evaluation Consistency
Fine-tuning a general-purpose large language model (LLM) for a specific domain or task has become a routine procedure for ordinary users. However, fine-tuning is known to remove the safety alignment features of the model, even when the fine-tuning data does not contain any harmful content. We consider this to be a critical failure mode of LLMs due to the widespread uptake of fine-tuning, combined with the benign nature of the "attack". Most well-intentioned developers are likely unaware that th…

SDD: Self-Degraded Defense against Malicious Fine-tuning
Zixuan Chen, Weikai Lu, Xin Lin, Ziqian Zeng
https://arxiv.org/abs/2507.21182 https://arxiv.org/pd…

SDD: Self-Degraded Defense against Malicious Fine-tuning
Open-source Large Language Models (LLMs) often employ safety alignment methods to resist harmful instructions. However, recent research shows that maliciously fine-tuning these LLMs on harmful data can easily bypass these safeguards. To counter this, we theoretically uncover why malicious fine-tuning succeeds and identify potential defense strategies. Building on the theoretical analysis, we introduce the Self-Degraded Defense (SDD) framework. SDD encourages LLMs to produce high-quality but irr…

This https://arxiv.org/abs/2506.02308 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

MINT: Multimodal Instruction Tuning with Multimodal Interaction Grouping
Recent advances in multimodal foundation models have achieved state-of-the-art performance across a range of tasks. These breakthroughs are largely driven by new pre-training paradigms that leverage large-scale, unlabeled multimodal data, followed by instruction fine-tuning on curated labeled datasets and high-quality prompts. While there is growing interest in scaling instruction fine-tuning to ever-larger datasets in both quantity and scale, our findings reveal that simply increasing the numb…

InfoSteer: Steering Information Utility in Language Model Post-Training
Chunyuan Deng, Ruidi Chang, Hanjie Chen
https://arxiv.org/abs/2507.05158 https://…

InfoSteer: Steering Information Utility in Language Model Post-Training
Recent advancements in language models (LMs) gradually ushered in an era where post-training is crucial. Yet, post-training approaches such as supervised fine-tuning (SFT) do not guarantee effective use of knowledge acquired during pretraining. We therefore present \ours, a lightweight method that encourages parametric information utilization in LMs during post-training. This is achieved via treating FFN layer as associate key-value memory, and promotes the use of stored memory vectors via forw…

This https://arxiv.org/abs/2505.03793 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

LENSLLM: Unveiling Fine-Tuning Dynamics for LLM Selection
The proliferation of open-sourced Large Language Models (LLMs) and diverse downstream tasks necessitates efficient model selection, given the impracticality of fine-tuning all candidates due to computational constraints. Despite the recent advances in LLM selection, a fundamental research question largely remains nascent: how can we model the dynamic behaviors of LLMs during fine-tuning, thereby enhancing our understanding of their generalization performance across diverse downstream tasks? In …

CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering
Hang Lv, Sheng Liang, Hao Wang, Hongchao Gu, Yaxiong Wu, Wei Guo, Defu Lian, Yong Liu, Enhong Chen
https://arxiv.org/abs/2507.04756

CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering
Personalized text generation has become crucial for adapting language models to diverse and evolving users' personal context across cultural, temporal, and contextual dimensions. While existing methods often rely on centralized fine-tuning or static preference alignment, they struggle to achieve real-time adaptation under resource constraints inherent to personal devices. This limitation creates a dilemma: large cloud-based models lack access to localized user-specific information, while small …

SecureT2I: No More Unauthorized Manipulation on AI Generated Images from Prompts
Xiaodong Wu, Xiangman Li, Qi Li, Jianbing Ni, Rongxing Lu
https://arxiv.org/abs/2507.03636

SecureT2I: No More Unauthorized Manipulation on AI Generated Images from Prompts
Text-guided image manipulation with diffusion models enables flexible and precise editing based on prompts, but raises ethical and copyright concerns due to potential unauthorized modifications. To address this, we propose SecureT2I, a secure framework designed to prevent unauthorized editing in diffusion-based generative models. SecureT2I is compatible with both general-purpose and domain-specific models and can be integrated via lightweight fine-tuning without architectural changes. We catego…

MOTIF: Modular Thinking via Reinforcement Fine-tuning in LLMs
Purbesh Mitra, Sennur Ulukus
https://arxiv.org/abs/2507.02851 https://a…

MOTIF: Modular Thinking via Reinforcement Fine-tuning in LLMs
Recent advancements in the reasoning capabilities of large language models (LLMs) show that employing group relative policy optimization (GRPO) algorithm for reinforcement learning (RL) training allows the models to use more thinking/reasoning tokens for generating better responses. However, LLMs can generate only a finite amount of tokens while maintaining attention to the previously generated tokens. This limit, also known as the context size of an LLM, is a bottleneck in LLM reasoning with a…

Knowledge-Aware Self-Correction in Language Models via Structured Memory Graphs
Swayamjit Saha
https://arxiv.org/abs/2507.04625 https://

Knowledge-Aware Self-Correction in Language Models via Structured Memory Graphs
Large Language Models (LLMs) are powerful yet prone to generating factual errors, commonly referred to as hallucinations. We present a lightweight, interpretable framework for knowledge-aware self-correction of LLM outputs using structured memory graphs based on RDF triples. Without retraining or fine-tuning, our method post-processes model outputs and corrects factual inconsistencies via external semantic memory. We demonstrate the approach using DistilGPT-2 and show promising results on simpl…