Searcharxiv⌕ Search

arXiv subjects

Kemou Li

Publications and source records attributed to Kemou Li.

16 recordsLinked to original sources
arXiv2026

RiSE: Residual Subspace Expert for Generalizable Text-Centric Image Forgery Localization

As AI-assisted image editing becomes increasingly prevalent, Text-Centric Image Forgery Localization (TFL) is essential for protecting trust in financial and legal records. However, we observe that existing TFL detectors predominantly rely on Full-Parameter Fine-Tuning (FPFT) and fail to generalize to unseen forgery types because FPFT can induce low-rank feature collapse and overfit training artifacts. Meanwhile, prior detectors capture diverse forgery traces with a single unified representation, making subtle traces entangled and difficult to learn. To address these challenges, we propose RiSE, a Residual Subspace Expert model that preserves pre-trained priors by freezing principal SVD components and adapting only the residual subspaces with a ViT-only segmentor to avoid feature collapse. RiSE introduces forgery-aware residual experts, where each expert is trained independently on a training subset with a particular forgery type, and we show that cross-domain localization performance improves consistently as more experts are added. At inference, we introduce Latent Perturbation Confidence (LPC), which selects the most confident expert by analytically computing the latent-perturbed output. LPC enables efficient confidence estimation with a single forward pass, avoiding repeated forward passes on perturbed samples and improving expert selection. By training on synthetic forgery data, RiSE with 1 expert outperforms prior methods on real-world forgery images by 20.4% while reducing training steps by 10 times. Scaling the number of experts to 9 yields a 35.8% gain. RiSE remains effective with DCT feature, even when the majority of the parameters is frozen with RGB-only pretraining. Code: https://github.com/KahimWong/RiSE.

cs.CV↗
arXiv2026

CABAL: Multi-Agent Simulacra for Tracing the Effects of Collusive Bidding in Peer Review

Recent reports during the AAAI-27 review cycle highlight the risk of reviewers coordinating bids for reciprocal assignment advantage. Prior work treats bidding, reviewer assignment, and review manipulation as separate stages, leaving the lifecycle effects of collusive bidding unclear. Real-world analysis is further constrained by typically unobservable collusive intent and the lack of counterfactuals for the same conference. Motivated by this gap, we introduce \alg, an end-to-end multi-agent simulacra framework for studying reviewer assignment integrity by holding the conference environment fixed and configuring LLM-driven reviewer agents with honest or collusive policies. We further develop an affinity-guided collusive bidding strategy that uses mutual reviewer-paper affinities to construct collusion rings and select target papers, producing expertise-consistent rather than arbitrarily targeted attacks. Controlled experiments show that collusive bidding more than doubles target-paper capture and that assigned colluders score target papers about two points higher than honest co-reviewers, while conference-wide effects remain comparatively modest. Evaluated bid-phase detectors provide only limited evidence of collusion: in a fixed-triplet detector stress test, native positive-bid graphs are confounded by benign affinity, while a Very-High-only diagnostic view enables precise but low-coverage local recovery.

cs.AI↗
arXiv2026

Preemptive LLM Unlearning against Forbidden Capability Acquisition via Gradient Sealing

Open-weight LLMs are released not only as fixed products but also as substrates for downstream fine-tuning. This openness, however, creates legal and ethical risks because users may misuse fine-tuning to instill illicit knowledge or enable hostile operations. Model providers therefore need apre-release defense against such acquisition, motivating the problem of preemptive unlearning. Unlike retrospective unlearning, which removes capabilities already present in a fixed model, preemptive unlearning seeks to prevent their acquisition under unseen attack data and future fine-tuning procedures. Despite its practical importance, this setting remains largely unexplored, presents distinct challenges, and is therefore the central focus of our work. We first verify that existing retrospective methods provide insufficient pre-release protection. Even when forbidden capabilities are suppressed in current outputs, forbidden-domain data can still induce gradients through internal pathways, enabling later acquisition. Motivated by this finding, we propose a gradient-sealing principle that blocks these pathways by pushing relevant pre-activations into the negative region, where ReLU-family activations exhibit zero or near-zero derivatives. Experiments across multiple LLM families demonstrate our stronger resistance to downstream acquisition than retrospective baselines, validating gradient sealing as an effective mechanism for pre-release protection.

cs.LG↗
arXiv2026

LLM Persona Unlearning

Pre-training equips large language models (LLMs) with a broad repertoire of behavioral patterns associated with roles, styles, values, and goals. Post-training teaches conditional enactment and makes a helpful Assistant the default, but it does not erase alternative modes from the weights; explicit prompts can therefore elicit personas that repeatedly shape judgment, language, and action. In open-weight settings, runtime controls can be removed, motivating persona unlearning: a weight-level edit that makes a designated persona difficult to elicit and enact on unseen contexts. We introduce PersonaUnlearnBench, a model-specific paired benchmark spanning six LLMs from three families and five personas, with aligned forget/retain sets, held-out instruction paraphrases, and four-axis evaluation. The benchmark shows that standard unlearning methods cannot reliably erase the target persona without sacrificing meaningful generation or general utility. We therefore propose PaCE, which compares target and desirable responses to the same questions to locate an internal behavior direction, then trains target-prompt states away from the target mode and toward the matched desirable response. Experiments show that PaCE consistently suppresses target personas with high response quality and useful counterpart behavior, at moderate utility cost. These results establish persona unlearning as a distinct behavior-level editing problem and a practical route toward persistent control of latent LLM response policies.

cs.CL↗
arXiv2026

TRACE: Trajectory Return Attribution and Contrastive Erasure for Multi-Turn Safety

Safety-aligned large language models (LLMs) often refuse a harmful request but comply once the same goal is spread over several turns. Preference objectives score whole responses to single prompts, so their training loss alone cannot control risk on unseen histories. Our analysis gives sufficient conditions under which suppression at supervised single-turn contexts yields a bound on multi-turn trajectory risk. The bound accounts for coverage, transfer slack, and leakage, and characterizes contraction relative to a base-policy risk budget evaluated on the trained policy's contexts. TRACE (Trajectory Return Attribution and Contrastive Erasure) turns this principle into a token-level objective. On the safe response, each token is weighted by the discounted return of a refusal-attributable advantage. The advantage compares a frozen reference model with its refusal-ablated copy, allowing earlier response tokens to receive credit from later refusal-related evidence. At high-gap positions on rejected responses, TRACE combines the observed token with policy-selected alternatives in the erasure target. A gradient-norm penalty replaces the retain set. Across five open-weight models and seven multi-turn attacks, TRACE gives the lowest attack success rate (ASR) in all 35 model and attack pairs, while the model utility evaluated on MMLU and HellaSwag drop by at most 1\.23 points. Source code can be found in the supplemental material.

cs.AI↗
arXiv2026

PACE: Provenance-Aware Capability Enforcement for Tool-Using LLM Agents

Tool-using large language model (LLM) agents turn generated text into real side effects, so poisoned tool metadata, retrieved pages, memory, and reusable skills can steer the next call. Vetting an artifact before admission does not settle this. A safe variant and a leaking variant can produce the same admission evidence, and a sound gate then cannot relax that site for either. We make that condition precise, which leaves the last boundary a deployment can still act on. We present Provenance-Aware Capability Enforcement (PACE), which mediates every tool call immediately before it executes. Path confinement proposes an executable cut of represented influence paths, while capability and effect verification checks schema-defined effects against authority compiled from the authenticated request. We distinguish the certified execution contract from the evaluated configuration, which can restore an authorized call after a proposed block or apply a declared repair. Confinement requires the final action to preserve the certified cut. On eight executable agent-security benchmarks with three target-model families, the evaluated configuration gives strictly lowest attack success in 62 of 79 eligible attack columns and ties in 14; full-benchmark native utility loses at most three points relative to the undefended agent. A complete ablation over 1167 paired cases attributes most security gains to effect verification and refusal control to boundary adaptation. A reduced-scale adaptive search succeeds on 0/30 out-of-authority targets against the defense.

cs.CR↗
arXiv2026

SEED: Simple ViT and Evolving Harness for Explainable Text Forgery Detection

AI-assisted image editing threatens trust in financial, legal, and identity records. The GenText-Forensics Challenge at ACM MM 2026 addresses this by requiring structured forensic reports, in which integrating detection, pixel-level localization, and natural language explanation for multilingual text-centric forgery images. We present SEED, a modular system with three components. First, a similarity-guided pipeline augments training with diverse synthetic forgeries. Second, a single ViT, built on DINOv3 with LoRA adaptation, jointly performs detection and pixel-level localization while preserving pre-trained priors with minimal trainable parameters. Third, an evolving harness takes the detector's predictions and generates a complete forensic report via an MLLM, iteratively improved through a proposer-evaluator loop optimizing report quality. SEED ranked 3rd in the GenText-Forensics Challenge. Code and data are available at https://github.com/KahimWong/GenText-Forensics-3rd-Place.

cs.CV↗
arXiv2026

When Modalities Fail to Tango: Conformal Backdoor Detection in Multimodal Contrastive Learning

Backdoor attacks in multimodal contrastive learning (MCL) have garnered growing attention in recent years, as many downstream tasks critically depend on pre-trained MCL models. Existing detection-based defenses predominantly rely on the CLIPScore metric, under the assumption that poisoned pairs exhibit lower semantic similarity between the image and the caption. However, we identify two critical flaws remaining in existing methods: (1) the substantial overlap between CLIPScore distributions of benign and poisoned pairs undermines the reliability of this metric, and (2) fixed-threshold detection cannot provide statistical guarantees for ambiguous samples within overlapping regions. To overcome these limitations, we propose integrating conformal prediction (CP), a statistical framework that quantifies uncertainty through nonconformity scores (NCSs), to establish provable confidence bounds for detecting poisoned image-caption pairs. Building on CP, we introduce CASCADE, a novel two-stage Coarse-to-Fine Conformal Backdoor Detection framework. The coarse-grained stage uses cross-modality consistency to identify high-confidence benign and poisoned pairs. In the fine-grained stage, a reference set is constructed from high-confidence poisoned pairs, and instance-level NCSs based on text-space similarity are computed for each sample in the unidentified subset. These NCSs measure conformity to the poisoning distribution and enable precise identification of latent poisoned pairs within the unidentified subset. Extensive experiments on the large-scale CC3M dataset demonstrate that CASCADE achieves an average FPR of 5.79% at 100% TPR and an average AUROC of 0.9867 across diverse attacks, while remaining effective against adaptive attacks.

cs.CR↗
arXiv2026

NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild

This paper presents an overview of the NTIRE 2026 Challenge on Robust AI-Generated Image Detection in the Wild, held in conjunction with the NTIRE workshop at CVPR 2026. The goal of this challenge was to develop detection models capable of distinguishing real images from generated ones in realistic scenarios: the images are often transformed (cropped, resized, compressed, blurred) for practical usage, and therefore, the detection models should be robust to such transformations. The challenge is based on a novel dataset consisting of 108,750 real and 185,750 AI-generated images from 42 generators comprising a large variety of open-source and closed-source models of various architectures, augmented with 36 image transformations. Methods were evaluated using ROC AUC on the full test set, including both transformed and untransformed images. A total of 511 participants registered, with 20 teams submitting valid final solutions. This report provides a comprehensive overview of the challenge, describes the proposed solutions, and can be used as a valuable reference for researchers and practitioners in increasing the robustness of the detection models to real-world transformations.

cs.CV↗
arXiv2026

$k$NNProxy: Efficient Training-Free Proxy Alignment for Black-Box Zero-Shot LLM-Generated Text Detection

LLM-generated text (LGT) detection is essential for reliable forensic analysis and for mitigating LLM misuse. Existing LGT detectors can generally be categorized into two broad classes: learning-based approaches and zero-shot methods. Compared with learning-based detectors, zero-shot methods are particularly promising because they eliminate the need to train task-specific classifiers. However, the reliability of zero-shot methods fundamentally relies on the assumption that an off-the-shelf proxy LLM is well aligned with the often unknown source LLM, a premise that rarely holds in real-world black-box scenarios. To address this discrepancy, existing proxy alignment methods typically rely on supervised fine-tuning of the proxy or repeated interactions with commercial APIs, thereby increasing deployment costs, exposing detectors to silent API changes, and limiting robustness under domain shift. Motivated by these limitations, we propose the $k$-nearest neighbor proxy ($k$NNProxy), a training-free and query-efficient proxy alignment framework that repurposes the $k$NN language model ($k$NN-LM) retrieval mechanism as a domain adapter for a fixed proxy LLM. Specifically, a lightweight datastore is constructed once from a target-reflective LGT corpus, either via fixed-budget querying or from existing datasets. During inference, nearest-neighbor evidence induces a token-level predictive distribution that is interpolated with the proxy output, yielding an aligned prediction without proxy fine-tuning or per-token API outputs. To improve robustness under domain shift, we extend $k$NNProxy into a mixture of proxies (MoP) that routes each input to a domain-specific datastore for domain-consistent retrieval. Extensive experiments demonstrate strong detection performance of our method.

cs.CL↗
arXiv2026

FeatDistill: A Feature Distillation Enhanced Multi-Expert Ensemble Framework for Robust AI-generated Image Detection

The rapid iteration and widespread dissemination of deepfake technology have posed severe challenges to information security, making robust and generalizable detection of AI-generated forged images increasingly important. In this paper, we propose FeatDistill, an AI-generated image detection framework that integrates feature distillation with a multi-expert ensemble, developed for the NTIRE Challenge on Robust AI-Generated Image Detection in the Wild. The framework explicitly targets three practical bottlenecks in real-world forensics: degradation interference, insufficient feature representation, and limited generalization. Concretely, we build a four-backbone Vision Transformer (ViT) ensemble composed of CLIP and SigLIP variants to capture complementary forensic cues. To improve data coverage, we expand the training set and introduce comprehensive degradation modeling, which exposes the detector to diverse quality variations and synthesis artifacts commonly encountered in unconstrained scenarios. We further adopt a two-stage training paradigm: the model is first optimized with a standard binary classification objective, then refined by dense feature-level self-distillation for representation alignment. This design effectively mitigates overfitting and enhances semantic consistency of learned features. At inference time, the final prediction is obtained by averaging the probabilities from four independently trained experts, yielding stable and reliable decisions across unseen generators and complex degradations. Despite the ensemble design, the framework remains efficient, requiring only about 10 GB peak GPU memory. Extensive evaluations in the NTIRE challenge setting demonstrate that FeatDistill achieves strong robustness and generalization under diverse ``in-the-wild'' conditions, offering an effective and practical solution for real-world deepfake image detection.

cs.CV↗
arXiv2026

LLM Unlearning with LLM Beliefs

Large language models trained on vast corpora inherently risk memorizing sensitive or harmful content, which may later resurface in their outputs. Prevailing unlearning methods generally rely on gradient ascent and its variants to lower the probability of specific target responses. However, we find that this strategy induces a critical side effect: probability mass is redistributed into high-likelihood regions, often corresponding to semantically related rephrasings of the targets. We refer to this as the squeezing effect, which explains why many methods yield merely spurious unlearning, a problem further obscured by automated metrics (e.g., ROUGE, truth ratio) that misreport actual success. To address this, we propose a bootstrapping (BS) framework that explicitly links the squeezing effect with the model's own high-confidence generations, namely its model beliefs. Since model beliefs inherently capture the very high-likelihood regions where probability mass is squeezed, incorporating them into the unlearning objective directly counters the squeezing effect. By jointly suppressing both target responses and model beliefs, BS-T (token) attenuates high-probability tokens, whereas BS-S (sequence) removes entire high-confidence generations, together achieving more thorough forgetting while preserving utility. Extensive experiments across diverse benchmarks with various model families confirm the effectiveness of our approach.

cs.LG↗
arXiv2026

AEGIS: Adversarial Target-Guided Retention-Data-Free Robust Concept Erasure from Diffusion Models

Concept erasure helps stop diffusion models (DMs) from generating harmful content; but current methods face robustness retention trade off. Robustness means the model fine-tuned by concept erasure methods resists reactivation of erased concepts, even under semantically related prompts. Retention means unrelated concepts are preserved so the model's overall utility stays intact. Both are critical for concept erasure in practice, yet addressing them simultaneously is challenging, as existing works typically improve one factor while sacrificing the other. Prior work typically strengthens one while degrading the other, e.g., mapping a single erased prompt to a fixed safe target leaves class level remnants exploitable by prompt attacks, whereas retention-oriented schemes underperform against adaptive adversaries. This paper introduces Adversarial Erasure with Gradient Informed Synergy (AEGIS), a retention-data-free framework that advances both robustness and retention.

cs.LG↗
arXiv2025

FontGuard: A Robust Font Watermarking Approach Leveraging Deep Font Knowledge

The proliferation of AI-generated content brings significant concerns on the forensic and security issues such as source tracing, copyright protection, etc, highlighting the need for effective watermarking technologies. Font-based text watermarking has emerged as an effective solution to embed information, which could ensure copyright, traceability, and compliance of the generated text content. Existing font watermarking methods usually neglect essential font knowledge, which leads to watermarked fonts of low quality and limited embedding capacity. These methods are also vulnerable to real-world distortions, low-resolution fonts, and inaccurate character segmentation. In this paper, we introduce FontGuard, a novel font watermarking model that harnesses the capabilities of font models and language-guided contrastive learning. Unlike previous methods that focus solely on the pixel-level alteration, FontGuard modifies fonts by altering hidden style features, resulting in better font quality upon watermark embedding. We also leverage the font manifold to increase the embedding capacity of our proposed method by generating substantial font variants closely resembling the original font. Furthermore, in the decoder, we employ an image-text contrastive learning to reconstruct the embedded bits, which can achieve desirable robustness against various real-world transmission distortions. FontGuard outperforms state-of-the-art methods by +5.4%, +7.4%, and +5.8% in decoding accuracy under synthetic, cross-media, and online social network distortions, respectively, while improving the visual quality by 52.7% in terms of LPIPS. Moreover, FontGuard uniquely allows the generation of watermarked fonts for unseen fonts without re-training the network. The code and dataset are available at https://github.com/KAHIMWONG/FontGuard.

cs.CV↗
arXiv2024

DAT: Improving Adversarial Robustness via Generative Amplitude Mix-up in Frequency Domain

To protect deep neural networks (DNNs) from adversarial attacks, adversarial training (AT) is developed by incorporating adversarial examples (AEs) into model training. Recent studies show that adversarial attacks disproportionately impact the patterns within the phase of the sample's frequency spectrum -- typically containing crucial semantic information -- more than those in the amplitude, resulting in the model's erroneous categorization of AEs. We find that, by mixing the amplitude of training samples' frequency spectrum with those of distractor images for AT, the model can be guided to focus on phase patterns unaffected by adversarial perturbations. As a result, the model's robustness can be improved. Unfortunately, it is still challenging to select appropriate distractor images, which should mix the amplitude without affecting the phase patterns. To this end, in this paper, we propose an optimized Adversarial Amplitude Generator (AAG) to achieve a better tradeoff between improving the model's robustness and retaining phase patterns. Based on this generator, together with an efficient AE production procedure, we design a new Dual Adversarial Training (DAT) strategy. Experiments on various datasets show that our proposed DAT leads to significantly improved robustness against diverse adversarial attacks.

cs.LG↗
arXiv2023

Regroup Median Loss for Combating Label Noise

The deep model training procedure requires large-scale datasets of annotated data. Due to the difficulty of annotating a large number of samples, label noise caused by incorrect annotations is inevitable, resulting in low model performance and poor model generalization. To combat label noise, current methods usually select clean samples based on the small-loss criterion and use these samples for training. Due to some noisy samples similar to clean ones, these small-loss criterion-based methods are still affected by label noise. To address this issue, in this work, we propose Regroup Median Loss (RML) to reduce the probability of selecting noisy samples and correct losses of noisy samples. RML randomly selects samples with the same label as the training samples based on a new loss processing method. Then, we combine the stable mean loss and the robust median loss through a proposed regrouping strategy to obtain robust loss estimation for noisy samples. To further improve the model performance against label noise, we propose a new sample selection strategy and build a semi-supervised method based on RML. Compared to state-of-the-art methods, for both the traditionally trained and semi-supervised models, RML achieves a significant improvement on synthetic and complex real-world datasets. The source code of the paper has been released.

cs.LG↗