LoRA-BAM: Input Filtering for Fine-tuned LLMs via Boxed Abstraction Monitors over LoRA Layers

• URL: http://arxiv.org/abs/2506.00998v1
• Date: Sun, 01 Jun 2025 12:58:32 GMT
• Title: LoRA-BAM: Input Filtering for Fine-tuned LLMs via Boxed Abstraction Monitors over LoRA Layers
• Abstract summary: Fine-tuning large language models (LLMs) improves performance on domain-specific tasks but can lead to overfitting, making them unreliable on out-of-distribution (OoD) queries.<n>We propose LoRA-BAM - a method that adds OoD detection monitors to the LoRA layer using boxed abstraction to filter questions beyond the model's competence.
• Score: 3.674863913115432
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Fine-tuning large language models (LLMs) improves performance on domain-specific tasks but can lead to overfitting, making them unreliable on out-of-distribution (OoD) queries. We propose LoRA-BAM - a method that adds OoD detection monitors to the LoRA layer using boxed abstraction to filter questions beyond the model's competence. Feature vectors from the fine-tuning data are extracted via the LLM and clustered. Clusters are enclosed in boxes; a question is flagged as OoD if its feature vector falls outside all boxes. To improve interpretability and robustness, we introduce a regularization loss during fine-tuning that encourages paraphrased questions to stay close in the feature space, and the enlargement of the decision boundary is based on the feature variance within a cluster. Our method complements existing defenses by providing lightweight and interpretable OoD detection.

Related papers

• OmniVL-Guard Pro: A Tool-Augmented Agent for Omnibus Vision-Language Forensics [63.13200245209719]
  We propose textbf OmniVL-Guard Pro, a tool-augmented agent that extends unified forensics from closed-world prediction to open-world clues-driven reasoning.<n>To generate high-quality tool-reasoning trajectories, we introduce textbfTree-Structured Self-Evolving Tool Trajectory Generation.<n>We also propose textbfChecker-Guided Agentic Reinforcement Learning, which provides process-level supervision to penalize cases where the answer is correct but the reasoning is distorted.
  arXiv  Detail & Related papers  (2026-05-16T12:26:04Z)
• Logic-Regularized Verifier Elicits Reasoning from LLMs [63.65875399266337]
  We propose LOVER, an unsupervised verifier regularized by logical rules.<n>Lover treats theverifier as a binary latent variable, utilizinginternal activations and enforcing three logical constraints.<n>Lover significantly outperforms unsupervised baselines.
  arXiv  Detail & Related papers  (2026-05-07T09:03:49Z)
• Exploring Data-Free LoRA Transferability for Video Diffusion Models [41.981710361053196]
  Video diffusion models leveraging step distillation or causal distillation have achieved remarkable performance.<n> adapting existing LoRAs to these variants remains a critical challenge due to weight space mismatches.<n>We propose Cluster-Aware Spectral Arbitration (CASA), a data-free framework that dynamically arbitrates between safeguarding the target's manifold and restoring LoRA alignment.
  arXiv  Detail & Related papers  (2026-05-03T15:22:53Z)
• GUIDED: Granular Understanding via Identification, Detection, and Discrimination for Fine-Grained Open-Vocabulary Object Detection [54.19989440021701]
  Fine-grained open-vocabulary object detection (FG-OVD) aims to detect novel object categories described by attribute-rich texts.<n>FG-OVD achieves new state-of-the-art results, demonstrating the benefits of disentangled modeling and modular optimization.
  arXiv  Detail & Related papers  (2026-03-27T22:08:11Z)
• Tensorized Clustered LoRA Merging for Multi-Task Interference [29.488565631016332]
  In multi-task settings, merging LoRA adapters trained on heterogeneous sources frequently causes textittask interference, degrading downstream performance.<n>We propose a tensorized clustered LoRA library targeting to address the task interference.<n> TC-LoRA achieves +1.4% accuracy on Phi-3 and +2.3% on Mistral-7B, demonstrating the effectiveness of TC-LoRA in LLM adaptation.
  arXiv  Detail & Related papers  (2025-08-06T01:26:43Z)
• AirLLM: Diffusion Policy-based Adaptive LoRA for Remote Fine-Tuning of LLM over the Air [14.089748643405498]
  AirLLM is a hierarchical diffusion policy framework for communication-aware LoRA adaptation.<n>AirLLM consistently enhances fine-tuning performance while significantly reducing transmission costs.
  arXiv  Detail & Related papers  (2025-07-15T17:36:37Z)
• Backdoor Cleaning without External Guidance in MLLM Fine-tuning [76.82121084745785]
  Believe Your Eyes (BYE) is a data filtering framework that leverages attention entropy patterns as self-supervised signals to identify and filter backdoor samples.<n>It achieves near-zero attack success rates while maintaining clean-task performance.
  arXiv  Detail & Related papers  (2025-05-22T17:11:58Z)
• Federated Sketching LoRA: On-Device Collaborative Fine-Tuning of Large Language Models [18.782733798668122]
  Fine-tuning large language models (LLMs) on devices remains a challenging problem.<n>Recent works have fused low-rank adaptation (LoRA) techniques with federated fine-tuning to mitigate challenges associated with device model sizes and data scarcity.<n>We propose federated sketching LoRA, which leverages a sketching mechanism to enable devices to selectively update submatrices of global LoRA modules maintained by the server.
  arXiv  Detail & Related papers  (2025-01-31T18:44:35Z)
• Provenance: A Light-weight Fact-checker for Retrieval Augmented LLM Generation Output [49.893971654861424]
  We present a light-weight approach for detecting nonfactual outputs from retrieval-augmented generation (RAG) We compute a factuality score that can be thresholded to yield a binary decision. Our experiments show high area under the ROC curve (AUC) across a wide range of relevant open source datasets.
  arXiv  Detail & Related papers  (2024-11-01T20:44:59Z)
• Sparse Semi-DETR: Sparse Learnable Queries for Semi-Supervised Object Detection [12.417754433715903]
  We introduce Sparse Semi-DETR, a novel transformer-based, end-to-end semi-supervised object detection solution. Sparse Semi-DETR incorporates a Query Refinement Module to enhance the quality of object queries, significantly improving detection capabilities for small and partially obscured objects. On the MS-COCO and Pascal VOC object detection benchmarks, Sparse Semi-DETR achieves a significant improvement over current state-of-the-art methods.
  arXiv  Detail & Related papers  (2024-04-02T10:22:23Z)
• CRaSh: Clustering, Removing, and Sharing Enhance Fine-tuning without Full Large Language Model [22.870512676002463]
  This paper focuses on Offsite-Tuning (OFT), a representative technique that transfers transformer blocks between centralized LLMs and downstream emulators. Inspired by these observations, we propose CRaSh, involving Clustering, Removing, and Sharing, a training-free strategy to derive improved emulators from LLMs. Our findings demonstrate a linear connectivity among these optima falling over the same basin, thereby highlighting the effectiveness of CRaSh and OFT.
  arXiv  Detail & Related papers  (2023-10-24T03:08:58Z)
• Amortizing intractable inference in large language models [56.92471123778389]
  We use amortized Bayesian inference to sample from intractable posterior distributions. We empirically demonstrate that this distribution-matching paradigm of LLM fine-tuning can serve as an effective alternative to maximum-likelihood training. As an important application, we interpret chain-of-thought reasoning as a latent variable modeling problem.
  arXiv  Detail & Related papers  (2023-10-06T16:36:08Z)
• Distribution-Aware Calibration for Object Detection with Noisy Bounding Boxes [58.2797274877934]
  We propose DIStribution-aware CalibratiOn (DISCO) to model the spatial distribution of proposals for calibrating supervision signals. Three distribution-aware techniques are developed to improve classification, localization, and interpretability.
  arXiv  Detail & Related papers  (2023-08-23T09:20:05Z)
• W2N:Switching From Weak Supervision to Noisy Supervision for Object Detection [64.10643170523414]
  We propose a novel WSOD framework with a new paradigm that switches from weak supervision to noisy supervision (W2N) In the localization adaptation module, we propose a regularization loss to reduce the proportion of discriminative parts in original pseudo ground-truths. Our W2N outperforms all existing pure WSOD methods and transfer learning methods.
  arXiv  Detail & Related papers  (2022-07-25T12:13:48Z)
• Boosting Weakly Supervised Object Detection via Learning Bounding Box Adjusters [76.36104006511684]
  Weakly-supervised object detection (WSOD) has emerged as an inspiring recent topic to avoid expensive instance-level object annotations. We defend the problem setting for improving localization performance by leveraging the bounding box regression knowledge from a well-annotated auxiliary dataset. Our method performs favorably against state-of-the-art WSOD methods and knowledge transfer model with similar problem setting.
  arXiv  Detail & Related papers  (2021-08-03T13:38:20Z)
• ANL: Anti-Noise Learning for Cross-Domain Person Re-Identification [25.035093667770052]
  We propose an Anti-Noise Learning (ANL) approach, which contains two modules. FDA module is designed to gather the id-related samples and disperse id-unrelated samples, through the camera-wise contrastive learning and adversarial adaptation. Reliable Sample Selection ( RSS) module utilizes an Auxiliary Model to correct noisy labels and select reliable samples for the Main Model.
  arXiv  Detail & Related papers  (2020-12-27T02:38:45Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.