Who Bridges Safety? Identifying and Targeting Cross-Lingual Shared Safety Pathways

Uncovering the internal mechanisms underlying the safety capabilities of large language models (LLMs) is crucial for developing trustworthy artificial intelligence. Currently, mechanistic interpretability studies on multilingual safety are largely confined to local components, such as isolated neurons. However, this static and fragmented perspective overlooks the synergy among components and fails to elucidate how safety signals dynamically propagate within the model to drive safety decisions ultimately. In this work, we move beyond isolated neurons to identify and target the cross-layer functional pathways formed during safety signal propagation, thereby uncovering the mechanisms driving the cross-lingual safety gap. Specifically, we first identify monolingual safety pathways and validate their impact on refusing harmful requests. Subsequent cross-lingual analyses reveal a sparse subset of cross-lingual shared safety pathways, confirming that this intersection acts as the internal bridge transferring safety capabilities from high-resource (HR) languages to non-high-resource (NHR) languages. Building on these mechanistic findings, we propose a pathways-targeted alignment method based on the cross-lingual shared safety pathways. Experimental results show that updating only a small fraction of pathway parameters significantly improves safety in NHR languages while largely preserving the model's general capabilities.

Hidden APIs in Language Models: Discovering Reusable Causal Interfaces from Forked Futures

Identical language-model answers can arise from hidden states that support different future computations, so current-answer probes do not establish a reusable internal interface. We introduce forked futures: future operations are sampled only after a prefix state has formed, and states are compared through the response distributions induced by those operations. This yields an empirical causal quotient over hidden states without requiring researcher-specified latent labels. Shared, Local, Mixture, and Distributed interfaces then compete under prequential causal description length subject to future-signature fidelity and matched capacity constraints. In the two detailed model evaluations, Shared has the lowest held-out description length, with gains of 0.216 nats on Qwen2.5-1.5B and 0.294 nats on Llama-3-8B, while maintaining tightly clustered mean future-signature distortion; a five-backbone sweep preserves the positive direction of Sharedness Gain. The figure-aligned transplantation analysis gives Shared the strongest joint target-correctness, locality, copy-preservation, and composite profile, and API-aligned paths mediate 0.749 of the target effect versus 0.150 for matched null paths. In the blind four-class model-organism test, 14/16 architectures are recovered, with one observed non-Shared to Shared error among 12 non-Shared organisms. These results support an economical reusable causal interface within the tested operation banks, while keeping the claim explicitly conditional on the candidate architectures, interventions, and held-out futures.

LaP-Forensics: Latent-Pixel Consistency Guided Multimodal Reasoning for Deepfake Detection

Recent generative models can produce images with few obvious visual artifacts, weakening detectors and explanations that rely only on surface appearance. We present LaP-Forensics, a multimodal framework that augments RGB semantics with reconstruction-based forensic evidence. A frozen Stable Diffusion DDIM inversion-reconstruction model provides a fixed reconstruction reference, and its residual map measures local compatibility with that reference. Independent projectors encode the RGB image and residual map before a structured Where-What-Why model predicts a textual analysis and an artifact mask.Supervised fine-tuning is followed by Group Relative Policy Optimization (GRPO), whose reward combines mask overlap with output-structure and evidence-reference terms. These text-side terms encourage the model to refer to the consistency map but do not constitute a verifier of free-form textual truth. A separate image-level head fuses RGB and DDIM-residual class features. Experiments show cross-generator detection on UniversalFakeDetect and competitive artifact localization on the official SynthScars benchmark. Controlled cue-construction, inversion-horizon, component, reward-term, and counterfactual analyses support the utility of the residual stream under the evaluated settings, while free-form textual faithfulness and reliability under post-processing remain open limitations.

Text as Partial Constraint: Core-Residual Alignment for Robust Vision-Language Learning

Vision-language alignment powers open-vocabulary recognition, retrieval, and LVLM grounding, yet natural captions are often underspecified, making similarity brittle and overly confident under paraphrase and omitted details. We aim to learn representations whose matching is stable across caption views and whose confidence reflects how strongly text constrains an image. We propose Text as Partial Constraint (TPC), a core-residual alignment framework that treats multi-view captions as incomplete supervision. It distills a consensus semantic core as the alignment target, learns a single-view core predictor for standard inference with one query, and explicitly discourages vision-language similarity from depending on the orthogonal unsaid residual. An uncertainty-aware contrastive objective further softens alignment when caption views disagree, reducing overconfident updates under weak language constraints. Across zero-shot recognition and adversarial robustness, TPC achieves 81.42/64.05 Top-1 clean/robust accuracy on ImageNet and 76.19/52.03 on an Avg-14 transfer suite, while improving LVLM transfer with 85.16 POPE F1 and 59.57 OKVQA accuracy under an LLaVA-1.5-7B stack. These results suggest that modeling text as a partial constraint is a practical and principled route to more reliable vision-language representations under underspecified language supervision.

Hidden Forgetting in Continual Multimodal Learning: When Accuracy Survives but Grounding Fails

Multimodal large language models must continually adapt to evolving tasks and domains, yet standard continual learning metrics mainly measure whether old answers remain correct, leaving the stability of multimodal grounding largely unexamined. We study this overlooked failure mode and ask whether a continually adapted MLLM can preserve not only what it answers, but also how it uses visual, textual, OCR, chart, and document evidence. We identify \emph{hidden evidence-use forgetting}, where answer accuracy is retained while the model silently shifts toward different or less grounded evidence channels, and propose \textsc{RCL}, a replay-free reliance-constrained continual learning framework. \textsc{RCL} freezes the previous checkpoint as a behavioral reference, estimates teacher and student evidence-reliance profiles through counterfactual channel interventions, and jointly optimizes task learning, prediction preservation, and reliance preservation without adding inference-time cost. Across CoIN, COAST, MCITlib, and an evidence-sensitive multimodal stream, \textsc{RCL} consistently improves final performance and reduces forgetting over replay-free, PEFT, routing, and memory-assisted baselines, while substantially lowering modality reliance drift, dominant evidence flips, and hidden forgetting rates. These results suggest that robust continual multimodal learning requires preserving the evidence path behind correct answers, not merely the answers themselves.

InduceKV: Fixed-Footprint Continual Adaptation of Multimodal LLMs via Inducing KV Memories

Multimodal large language models must adapt to evolving tasks and domains, yet continual improvement under bounded deployment footprint remains difficult because repeated parameter updates or growing replay stores can accumulate adaptation state over time. We study fixed-footprint continual adaptation: the deployed adaptation state is kept under a fixed memory budget, while the backbone model is left unchanged and task-specific updates are externalized. We propose InduceKV, a retrieval-based method that stores each selected training prefix as an attention-ready memory entry, consisting of a frozen retrieval key and compact layerwise key--value (KV) payloads that can be appended to the model's self-attention cache. Under a strict memory budget, InduceKV constructs a compact inducing set through bilevel selection: a lightweight calibration is fit for retrieval, while the selected memory balances current-task likelihood, anchor-based retention, and coverage in the frozen retrieval space. Across task-incremental instruction tuning, continual VQA, domain-incremental adaptation, and lifelong multimodal instruction tuning, InduceKV consistently improves over PEFT, MoE, replay, and prompt-retrieval baselines under matched memory budgets. We further report backbone-matched, stage-1 CoIN, compute-matched, and scalability diagnostics, showing that the gains are not due to a stronger backbone, replay alone, or an unbounded candidate pool.

HG-Bench: A Benchmark for Multi-Page Handwritten Answer-Region Grounding in Automated Homework Assessment

Automated homework assessment depends not only on recognizing student answers, but also on accurately locating where each answer and each intermediate reasoning step appears in noisy, multi-page handwritten work. This paper addresses the missing evaluation setting of page-aware, two-level answer-region grounding: given a sequence of homework page images, a model must localize complete answer regions and their ordered step-level subregions. We introduce HG-Bench, a benchmark of 500 human-annotated K-12 homework samples curated from a 1,489,278-image source pool, with question-level and step-level boxes linked by a hierarchical containment constraint. HG-Bench is paired with a page-aware evaluation protocol that separately measures complete-answer localization (FA) and step-level decomposition (FSm), revealing whether models truly ground the spatial structure of student reasoning rather than merely parse visible text. Across frontier closed-source APIs and competitive open-weight VLMs, no zero-shot system exceeds 55.22% on FA or 48.22% on FSm, while a GLM-4.6V 9B reference model fine-tuned on ~10k in-domain examples reaches 74.97/72.26. These results identify step-level handwritten grounding as a concrete capability gap and provide a reproducible benchmark, evaluation protocol, and trained reference point for future work on automated homework assessment.

Attention-Spectrum Regularization for Replay-Free Continual Multimodal LLMs

Multimodal large language models (MLLMs) are increasingly required to adapt to non-stationary streams of visual domains, question types, and user instructions, yet continual fine-tuning often causes severe forgetting of previously acquired multimodal skills. Existing continual vision-language methods mainly preserve outputs, replay data or pseudo-data, regularize embedding geometry, or allocate task-specific parameters, but they provide limited control over how internal cross-modal attention patterns supporting old skills drift during adaptation. We propose Attention-Spectrum Regularization (ASR), a replay-free continual learning framework that preserves skill-conditioned structures of cross-modal attention. ASR treats cross-attention maps as two-dimensional signals, summarizes their scale and directional properties into compact spectral statistics, and stores only skill-wise prototype distributions instead of replaying past image-question pairs, generated pseudo-examples, or old-stage teacher snapshots. In later stages, a phase-invariant spectral regularizer constrains harmful drift of these prototypes while allowing instance-level attention to adapt to new tasks. We provide theoretical analysis showing that skill-conditioned spectral drift controls forgetting under a spectral sufficiency assumption, and that Fourier power spectra are stable to spatial translations and bounded perturbations. Experiments on continual VQA and multimodal instruction-tuning benchmarks, including VQA v2, VQACL, CLT-VQA, CoIN, and UCIT, show that ASR consistently improves final performance and reduces forgetting over strong replay-, regularization-, and adapter-based baselines. Preserving skill-level attention structure is an effective and lightweight mechanism for continual MLLMs. Code is available at https://github.com/Creative-zcx/attention-spectrum-replay

Prototype-Aligned Federated Soft-Prompts for Continual Web Personalization

Continual web personalization is essential for engagement, yet real-world non-stationarity and privacy constraints make it hard to adapt quickly without forgetting long-term preferences. We target this gap by seeking a privacy-conscious, parameter-efficient interface that controls stability-plasticity at the user/session level while tying user memory to a shared semantic prior. We propose ProtoFed-SP, a prompt-based framework that injects dual-timescale soft prompts into a frozen backbone: a fast, sparse short-term prompt tracks session intent, while a slow long-term prompt is anchored to a small server-side prototype library that is continually refreshed via differentially private federated aggregation. Queries are routed to Top-M prototypes to compose a personalized prompt. Across eight benchmarks, ProtoFed-SP improves NDCG@10 by +2.9% and HR@10 by +2.0% over the strongest baselines, with notable gains on Amazon-Books (+5.0% NDCG vs. INFER), H&M (+2.5% vs. Dual-LoRA), and Taobao (+2.2% vs. FedRAP). It also lowers forgetting (AF) and Steps-to-95% and preserves accuracy under practical DP budgets. Our contribution is a unifying, privacy-aware prompting interface with prototype anchoring that delivers robust continual personalization and offers a transparent, controllable mechanism to balance stability and plasticity in deployment.

Seeing through the Conflict: Transparent Knowledge Conflict Handling in Retrieval-Augmented Generation

Large language models (LLMs) equipped with retrieval--the Retrieval-Augmented Generation (RAG) paradigm--should combine their parametric knowledge with external evidence, yet in practice they often hallucinate, over-trust noisy snippets, or ignore vital context. We introduce TCR (Transparent Conflict Resolution), a plug-and-play framework that makes this decision process observable and controllable. TCR (i) disentangles semantic match and factual consistency via dual contrastive encoders, (ii) estimates self-answerability to gauge confidence in internal memory, and (iii) feeds the three scalar signals to the generator through a lightweight soft-prompt with SNR-based weighting. Across seven benchmarks TCR improves conflict detection (+5-18 F1), raises knowledge-gap recovery by +21.4 pp and cuts misleading-context overrides by -29.3 pp, while adding only 0.3% parameters. The signals align with human judgements and expose temporal decision patterns.