论文检索

Social Aspects · Security

Ziyuan Fang, Tianyi Wei, Guanjie Wang, Weiming Zhang, Nenghai Yu, Wenbo Zhou

Recent advances in video generation have enabled highly realistic synthetic content, raising concerns about the integrity of digital media and motivating the development of benchmarks and detection methods for generated videos. Prior works have largely prioritized bolstering model generalization against unseen generators. However, we uncover a neglected factor: the quality distribution of real videos plays a pivotal role. Current training protocols suffer from a clear quality bias between real and fake data, prone to shortcut learning. Compounded by testing on similar real data distributions, this creates an illusion of generalization. In reality, these models fail to generalize when exposed to real data with significantly different quality profiles. To address this, we propose training with quality-matched real and fake data to mitigate bias. Building on this, we introduce a data expansion strategy that broadens the training set to comprehensively cover the full quality spectrum. This approach enables the model to learn quality-agnostic features for detection, thereby achieving generalization across real data of varying qualities and enhancing real-world applicability. Extensive experiments demonstrate that our method scales well across diverse backbones, consistently enhancing the generalization capability of existing models.

Social Aspects · Security

Hongyi Zhou, Jianfeng Pan, Min Peng, Shaomang Huang, Xuling Zhang

Endpoint Detection and Response (EDR) systems are crucial for identifying malicious activities on endpoint devices, yet existing methods struggle to efficiently model ultra-long log sequences and to provide interpretable reasoning for security analysts. We propose WatchLog, a novel framework that represents raw logs as video-structured data, enabling scalable and expressive video-language modeling of endpoint behaviors. Each event is encoded as a key–value-guided image, and the resulting images are temporally organized into a video sequence. To capture long-range dependencies, WatchLog employs a temporal cross-attention adapter that enables pixel-wise interaction across time. The adapter acts as an auxiliary temporal reasoning pathway, aligning spatial representations with relevant temporal contexts while preserving the original behavioral semantics. We adopt a two-stage pre-training strategy followed by supervised fine-tuning to generate behavior explanations grounded in event-level semantics and detection outcomes. Experiments on our newly constructed EDR8M-20R dataset and a public benchmark demonstrate that WatchLog consistently outperforms state-of-the-art methods in detection accuracy and recall, while offering more interpretable reasoning traces and significantly improved inference efficiency. Extensive ablation studies further support the robustness and interpretability of the proposed method.

Social Aspects · Security

Chengran Yang, zichao wei, Heminghao Deng, Jinfeng Jiang, Zhensu Sun, Ting Zhang, Tianyi Wu, Ming Wen, David Lo

Large Language Model (LLM) based code generation is predominantly formulated as a strictly monotonic process, appending tokens linearly to an immutable prefix. This formulation contrasts to the cognitive process of programming, which is inherently interleaved with forward generation and on-the-fly revision. While prior works attempt to introduce revision via post-hoc agents or external static tools, they either suffer from high latency or fail to leverage the model's intrinsic semantic reasoning. In this paper, we propose Stream of Revision, a paradigm shift that elevates code generation from a monotonic stream to a dynamic, self-correcting trajectory by leveraging model's intrinsic capabilities. We introduce specific action tokens that enable the model to seamlessly backtrack and edit its own history within a single forward pass. By internalizing the revision loop, our framework Stream of Revision allows the model to activate its latent capabilities just-in-time without external dependencies. Empirical results on secure code generation show that Stream of Revision significantly reduces vulnerabilities with minimal inference overhead. We release our code and dataset at \url{https://anonymous.4open.science/r/Stream-of-Revision-CB74/}, and models will be released upon publication.

Social Aspects · Security

Nikil Selvam, Jay Baxter, Sophie Hilgard, Brad Miller, Keith Coleman, Ellen Vitercik, Sanmi Koyejo

Crowdsourced fact-checking systems have been widely adopted by major social media companies such as X, Meta, Tiktok, and Google with the aim of combating misinformation at scale without relying on centralized editorial control. These systems have been developed around a common underlying algorithm: a bridging mechanism—based on matrix factorization—that surfaces notes indicating misinformation only when they receive support from diverse ideological groups rather than simple majority support. Although this algorithm is designed to be robust against traditional brigading, we demonstrate an attack showing that coordinated users can strategically fabricate diverse agreement in the system’s latent space to manipulate the scoring algorithm. We validate this attack on real-world production data and find that a surprisingly large number of notes’ scores can potentially be manipulated with a small number (< 10) of coordinated votes, raising the risk that adversaries could surface arbitrary notes on these social media platforms. We complement these findings with a theoretical analysis of voting strategies that surface arbitrary notes, revealing counterintuitive properties of the system: for instance, rating a note as “Not Helpful” can increase its helpfulness score. Finally, we develop a cost model quantifying manipulation effort and discuss potential mitigations. Following a responsible disclosure process, X's Community Notes team acknowledged this attack and has deployed mitigations based on our findings. We hope this work spurs further research into the robustness of crowdsourced fact-checking systems and, more broadly, bridging-based consensus mechanisms.

Social Aspects · Security

Jiazhen Yan, Ziqiang Li, Fan Wang, Boyu Wang, Ziwen He, Zhangjie Fu

The rapid progress of generative models such as GANs and diffusion models has led to the widespread proliferation of AI-generated images, raising concerns about misinformation, privacy violations, and trust erosion in digital media. Although large-scale multimodal models like CLIP offer strong transferable representations for detecting synthetic content, fine-tuning them often induces catastrophic forgetting, which degrades pre-trained priors and limits cross-domain generalization. To address this issue, we propose the Distillation-guided Gradient Surgery Network (DGS-Net), a novel framework that preserves transferable pre-trained priors while suppressing task-irrelevant components. Specifically, we introduce a gradient-space decomposition that separates harmful and beneficial descent directions during optimization. By projecting task gradients onto the orthogonal complement of harmful directions and aligning with beneficial ones distilled from a frozen CLIP encoder, DGS-Net achieves unified optimization of prior preservation and irrelevant suppression. Extensive experiments on 50 generative models demonstrate that our method outperforms state-of-the-art approaches by an average margin of 6.6%, achieving superior detection performance and generalization across diverse generation techniques.

Social Aspects · Safety

Xuyang Zhong, Qizhang Li, Yiwen Guo, Chen Liu

We propose **DualOptim+**, a novel optimization framework for improving machine unlearning in large language models. It introduces a base state to capture common representations shared by forgetting and retaining objectives and delta states to preserve objective-specific residuals. This architecture allows the optimizer to adaptively bridge shared and decoupled states based on the directional conflict between forgetting and retaining gradients. We further introduce DualOptim+ 8bit, a quantized variant that reduces memory overhead without compromising performance. Extensive experiments across fictitious, real-world, and safety alignment tasks demonstrate that DualOptim+ consistently achieves a superior trade-off between forgetting efficacy and model utility.

Social Aspects · Safety

Maya Okawa, Bo Zhao, Eric Bigelow, Rose Yu, Tomer Ullman, Ekdeep Singh Lubana, Hidenori Tanaka

As large language models (LLMs) increasingly power conversational agents, understanding how they model users' emotional states is critical for ethical deployment. Inspired by emotion wheels, i.e., a psychological framework that argues emotions organize hierarchically, we analyze probabilistic dependencies between emotional states in model outputs. We find that LLMs naturally form hierarchical emotion trees that align with human psychological models, and larger models develop more complex hierarchies. We also uncover systematic biases in emotion recognition across socioeconomic personas, with compounding misclassifications for intersectional, underrepresented groups. Human studies reveal striking parallels, suggesting that LLMs internalize aspects of social perception. Beyond highlighting emergent emotional reasoning in LLMs, our results hint at the potential of using cognitively-grounded theories for developing better model evaluations.

Social Aspects · Safety

Xinbo Wu, Huan Zhang, Abhishek Umrawal, Lav Varshney

As large language models grow increasingly capable, concerns about their safe deployment have intensified. While numerous alignment strategies aim to restrict harmful behavior, these defenses can still be circumvented through carefully designed adversarial prompts. In this work, we introduce a theoretical framework that formalizes a game between an attacker and a defender. Within this framework, we design a theoretical best-response attack strategy and show that it is closely related to many existing adversarial prompting methods. We further analyze the resulting game, characterize its equilibria, and reveal inherent advantages for the attacker. Drawing on our theoretical analysis, we also derive a provably optimal defense strategy. Empirically, we evaluate a practical instantiation of the theoretically optimal attack and observe stronger performance relative to existing adversarial prompting approaches in diverse settings encompassing different LLMs and benchmarks.

Reinforcement Learning · Deep RL

Gwangpyo Yoo, Woo Kyung Kim, Honguk Woo

In this work, we extend distributional reinforcement learning (RL) to develop a risk-sensitive multi-objective RL framework, with applications to domains such as finance and robotics. We achieve this by adopting vector-risk measures and approximating them via Knothe-Rosenblatt (KR) quantile regression. This approach directly extends the IQN framework to the multi-objective setting, aligns with the axiomatic definition of vector-risk measures, and guarantees that critics converge under the distributional Bellman operator. To mitigate the artificial ordering imposed by the KR map, we employ a transformer architecture without positional encoding, and introduce MO-TQC for training stability. We demonstrate improved performance on MO-Gymnasium benchmarks and use our framework to study risk-sensitive policies in multi-objective tasks.

Social Aspects · Safety

Nanxiang Jiang, Zhaoxin Fan, Baisen Wang, Daiheng Gao, Junhang Cheng, Jifeng Guo, Yalan Qin, Yeying Jin, Hongwei Zheng, Faguo Wu 等

Concept erasure serves as a vital safety mechanism for removing unwanted concepts from text-to-image (T2I) models. While extensively studied in U-Net and dual-stream architectures (e.g., Flux), this task remains under-explored in the recent emerging paradigm of single-stream diffusion transformers (e.g., Z-Image). In this new paradigm, text and image tokens are processed as a single unified sequence via shared parameters. Consequently, directly applying prior erasure methods typically leads to generation collapse. To bridge this gap, we introduce Z-Erase, the first concept erasure method tailored for single-stream T2I models. To guarantee stable image generation, Z-Erase first proposes a Stream Disentangled Concept Erasure Framework that decouples updates and enables existing methods on single-stream models. Subsequently, within this framework, we introduce Lagrangian-Guided Adaptive Erasure Modulation, a constrained algorithm that further balances the sensitive erasure-preservation trade-off. Moreover, we provide a rigorous convergence analysis proving that Z-Erase can converge to a Pareto stationary point. Experiments demonstrate that Z-Erase successfully overcomes the generation collapse issue, achieving state-of-the-art performance across a wide range of tasks.

Social Aspects · Safety

Aradhye Agarwal, Gurdit Singh Siyan, Yash Pandya, Joykirat Singh, Akshay Nambi, Ahmed Awadallah

Agentic language models operate in a fundamentally different safety regime than chat models: they must plan, call tools, and execute long-horizon actions where a single misstep, such as accessing files or entering credentials, can cause irreversible harm. Existing alignment methods, largely optimized for static generation and task completion, break down in these settings due to sequential decision-making, adversarial tool feedback, and overconfident intermediate reasoning. We introduce MOSAIC, a post-training framework that aligns agents for safe multi-step tool use by making safety decisions explicit and learnable. MOSAIC structures inference as a plan, check, then act or refuse loop, with explicit safety reasoning and refusal as first-class actions. To train without trajectory-level labels, we use preference-based reinforcement learning with pairwise trajectory comparisons, which captures safety distinctions often missed by scalar rewards. We evaluate MOSAIC zero-shot across three model families, Qwen2.5-7B, Qwen3-4B-Thinking, and Phi-4, and across out-of-distribution benchmarks spanning harmful tasks, prompt injection, benign tool use, and cross-domain privacy leakage. MOSAIC reduces harmful behavior by up to 50\%, increases harmful-task refusal by over 20\% on injection attacks, cuts privacy leakage, and preserves or improves benign task performance, demonstrating robust generalization across models, domains, and agentic settings.

Social Aspects · Safety

William Overman, Mohsen Bayati

As increasingly capable agents are deployed, a central safety challenge is how to retain meaningful human control without modifying the underlying system. We study a minimal control interface in which an agent chooses whether to act autonomously (play) or defer (ask), while a human simultaneously chooses whether to be permissive (trust) or engage in oversight (oversee), and model this interaction as a two-player Markov Game. When this game forms a Markov Potential Game, we prove an alignment guarantee: any increase in the agent's utility from acting more autonomously cannot decrease the human's value. This establishes a form of intrinsic alignment where the agent's incentive to seek autonomy is structurally coupled to the human's welfare. Practically, the framework induces a transparent control layer encouraging the agent to defer when risky and act when safe. While we use gridworld simulations to illustrate the emergence of this collaboration, our primary validation involves an agentic tool-use task where two 30B-parameter language models are fine-tuned via independent policy gradient. We demonstrate that even as the agents learn to coordinate on the fly, this framework effectively reduces safety violations in realistic, open-ended environments.

Kunlun Xu, YanQin Zhang, Wenwen Qiang, Jiahuan Zhou

Multimodal Continual Instruction Tuning (MCIT) aims to exploit the incrementally accumulated knowledge to process multimodal inputs of diverse tasks, where parameter routing is an important technology. Existing advanced methods typically rely on sample to task center similarity and cross-modal fusion with equal weight during routing. However, such solutions face two fundamental flaws: (1) Within each modality, sample to task center distance is sub-optimal for routing since the abundant intra-task diversity information is underleveraged. (2) Different modalities exhibit varying reliability across tasks, where the modality with inter-task ambiguity can easily misguide the routing result. To address these problems, we propose Hyperbolic Uncertainty-aware Modality-Balanced Routing (Hyper-LLaVA) to improve parameter routing capacity based on cross-modality task feature uncertainty modeling. Specifically, to improve intra-modality task matching, Hyper-LLaVA accesses the sample to task distribution similarity in the Hyperbolic space. Besides, to alleviate the degradation brought by unreliable modality, Hyper-LLaVA quantifies the task matching ambiguity within each modality to achieve adaptive balancing between task matching across modalities. Based on the complementary intra- and inter-modality task matching enhancement, our Hyper-LLaVA outperforms state-of-the-art approaches by large margins.

Social Aspects · Safety

Charles Wang, Keir Dorchen, Peter Jin

Contemporary AI safety is increasingly a full-stack discipline. It spans pretraining interventions, post-training alignment (instruction tuning, RLHF and preference-optimization variants), and deployment-time controls (guardrails, monitoring, and red-teaming). This paper argues these efforts optimize the wrong primary target when it comes to self-improving agents: behavioral compliance today rather than teachability tomorrow. Building on the concept of the utility-learning tension formalized by Wang et al., we argue that utility-driven self-modification can erode learnability itself, yielding structural incorrigibility as an emergent consequence of optimization. We therefore call for a shift in priorities from behavioral alignment to enforceable learnability floors that preserve long-run corrigibility under bounded intervention.

Social Aspects · Alignment

Xiao Lin, Zhining Liu, Ze Yang, Gaotang Li, Ruizhong Qiu, Shuke Wang, Hui Liu, Haotian Li, Yuchen Yan, Sumit Keswani 等

Recently, vision-language models have demonstrated increasing influence in morally sensitive domains such as autonomous driving and medical analysis, owing to their powerful multimodal reasoning capabilities. As these models are deployed in high-stakes real-world applications, it is of paramount importance to ensure that their outputs align with human moral values and remain within moral boundaries. However, existing work on moral alignment either focuses solely on textual modalities or relies heavily on AI-generated images, leading to distributional biases and reduced realism. To overcome these limitations, we introduce MORALISE, a comprehensive benchmark for evaluating the \underline{mor}al \underline{al}ignment of v\underline{is}ion-languag\underline{e} models (VLMs) using diverse, expert-verified real-world data. We begin by proposing a comprehensive taxonomy of 13 moral topics grounded in Turiel's Domain Theory, spanning the personal, interpersonal, and societal moral domains encountered in everyday life. Built on this framework, we manually curate 2,481 high-quality image-text pairs, each annotated with two fine-grained labels: (1) \textit{topic annotation}, identifying the violated moral topic(s), and (2) \textit{modality annotation}, indicating whether the violation arises from the image or the text. For evaluation, we encompass two tasks, \textit{moral judgment} and \textit{moral norm attribution}, to assess models' awareness of moral violations and their reasoning ability on morally salient content. Extensive experiments on 19 popular open- and closed-source VLMs show that MORALISE poses a significant challenge, revealing persistent moral limitations in current state-of-the-art models.

Social Aspects · Alignment

Cameron Tice, Puria Radmard, Samuel Ratnam, Andy Kim, David Africa, Kyle O&amp;#x27;Brien

Pretraining corpora contain extensive discourse about AI systems, yet the causal influence of this discourse on downstream alignment remains poorly understood. If prevailing descriptions of AI behaviour are predominantly negative, LLMs may internalise corresponding behavioural priors, giving rise to self-fulfilling misalignment. This paper provides the first controlled study of this hypothesis by pretraining 6.9B-parameter LLMs with varying amounts of (mis)alignment discourse. We find that discussion of AI contributes to misalignment. Upsampling synthetic training documents about AI misalignment leads to a notable increase in misaligned behaviour. Conversely, upsampling documents about aligned behaviour reduces misalignment scores from 45% to 9%. We consider this evidence of self-fulfilling alignment. These effects are dampened, but persist through post-training. Our findings establish the study of how pretraining data shapes alignment priors, or alignment pretraining, as a complement to post-training. We recommend practitioners pretrain for alignment as well as capabilities.

Social Aspects · Alignment

Cristina Garbacea

This position paper argues that \textbf{large language models should transition from learning aggregated human preferences to learning personalized, individual preferences}. Current approaches to training language models with reinforcement learning from human feedback (RLHF) aggregate diverse human preferences into singular reward models, fundamentally limiting their ability to serve heterogeneous user populations. This aggregation masks critical information about preference diversity, individual values, and contextual dependencies, effectively optimizing models for a hypothetical ``average user'' who may not exist. We critically examine these limitations, analyze the rich structure that human preferences encode, and make the case for personalized and adaptive language model systems. While personalization offers substantial benefits for diverse user populations, it also introduces serious safety risks including manipulation, filter bubbles, and value lock-in. We discuss these risks in depth, present alternative views and counterarguments to our position, and propose a concrete call to action for responsible development of preference-aware models that respect both individual autonomy and collective safety.

Social Aspects · Alignment

Yaxuan Kong, Hoyoung Lee, Yoontae Hwang, Alejandro Lopez-Lira, Bradford Levy, Dhagash Mehta, Qingsong Wen, CHANYEOL CHOI, Yongjae Lee, Stefan Zohren

Large Language Models (LLMs) are increasingly integrated into financial workflows, but evaluation practice has not kept up. Finance-specific biases can inflate performance, contaminate backtests, and make reported results useless for any deployment claim. We identify five recurring biases in financial LLM applications. They include look-ahead bias, survivorship bias, narrative bias, objective bias, and cost bias. These biases break financial tasks in distinct ways and they often compound to create an illusion of validity. We reviewed 164 papers from 2023 to 2025 and found that no single bias is discussed in more than 28 percent of studies. This position paper argues that **bias in financial LLM systems requires explicit attention and that structural validity should be enforced before any result is used to support a deployment claim.** We propose a Structural Validity Framework and an evaluation checklist with minimal requirements for bias diagnosis and future system design. The material is available at https://anonymous.4open.science/r/Fin-LLM-Checklists-8557/.

Applications · Computer Vision

Maosheng Su, Junlei Cao, Zhiyuan Wang, Shuo Wang, Ning Zhang, Jun Luo

Reinforcement learning-based Multimodal Large Language Models (MLLMs) provide new perspectives for visual grounding, yet face significant challenges in Camouflaged Object Detection (COD) where objects blend seamlessly with backgrounds. This stems primarily from: difficulties in multi-object matching, the detrimental effects of low-quality samples, and erroneously localizing visual distractors with similar textures to true objects. We propose Progressive Matching and Semantic-aware Policy Optimization (PMSPO), a curriculum learning-based framework that employs Sinkhorn multi-object matching IoU reward during training for multi-object alignment, utilizes Positive Learning Gain Filtering (PLGF) to curate high-quality samples, and transforms deep visual features into semantic contrastive reward rules to calibrate target background semantics. Experiments on COD benchmarks demonstrate that PMSPO achieves state-of-the-art (SOTA) performance among reinforcement learning methods across all evaluation metrics.

Theory · Learning Theory

Xutao Ma, Yixiao Huang, Hanlin Zhu, Somayeh Sojoudi

Autoregressive large language models (LLMs) have achieved remarkable success in many complex tasks, yet they can still fail in very simple logical reasoning such as the "reversal curse" --- when trained on forward knowledge data of the form "$A \rightarrow B$" (e.g., Alice's husband is Bob), the model is unable to deduce the reversal knowledge "$B \leftarrow A$" (e.g., Bob's wife is Alice) during test. Extensive prior research suggests that this failure is an inherent, fundamental limit of autoregressive causal LLMs, indicating that these models tend to memorize factual-level knowledge rather than capture higher-level rules. In this paper, we challenge this view by showing that this seemingly fundamental limit can be mitigated by slightly tweaking the training data with a simple regularization data recipe called the Identity Bridge of the form "$A \to A$" (e.g., The name of Alice is Alice). Theoretically, we prove that under this recipe, even a one-layer transformer can break the reversal curse by analyzing the implicit bias of gradient descent. Empirically, we show that a 1B pretrained language model finetuned with the proposed data recipe achieves a 40\% success rate on reversal tasks, in stark contrast to a near-zero success rate when trained solely on forward-knowledge data. Our work provides a novel theoretical foundation for the reversal curse and offers a principled, low-cost path to encouraging LLMs to learn higher-level rules from data.