alignment.html 关键词 alignment 的长期追踪 RSS，汇总历史命中文献。 zh-CN Wed, 22 Apr 2026 03:37:20 +0000 ../papers/arxiv-d363006cb185.html https://arxiv.org/abs/2604.19457v1#2026-04-22#alignment Wed, 22 Apr 2026 11:37:03 +0800 Long-horizon enterprise agents make high-stakes decisions (loan underwriting, claims adjudication, clinical review, prior authorization) under lossy memory, multi-step reasoning, and binding regulatory constraints. Current evaluation reports a single task-success scalar that conflates distinct failure modes and hides whether an agent is aligned with the standards its deployment environment requires. We propose that long-horizon decision behavior decomposes into four orthogonal alignment axes, e… ../papers/arxiv-3ca660d54bb4.html https://arxiv.org/abs/2604.19548v1#2026-04-22#alignment Wed, 22 Apr 2026 11:37:03 +0800 Large Language Model agents have rapidly evolved from static text generators into dynamic systems capable of executing complex autonomous workflows. To enhance reliability, multi-agent frameworks assigning specialized roles are increasingly adopted to enable self-reflection and mutual auditing. While such role-playing effectively leverages domain expert knowledge, we find it simultaneously induces a human-like cognitive bias known as Actor-Observer Asymmetry (AOA). Specifically, an agent acting… ../papers/arxiv-0d90d26515bd.html https://arxiv.org/abs/2604.19716v1#2026-04-22#alignment Wed, 22 Apr 2026 11:37:03 +0800 Large Language Models (LLMs) still struggle with multi-step logical reasoning. Existing approaches either purely refine the reasoning chain in natural language form or attach a symbolic solver as an external module. In this work, we instead ask whether LLMs contain a shared internal logical subspace that simultaneously aligns natural-language and symbolic-language views of the reasoning process. Our hypothesis is that this logical subspace captures logical reasoning capabilities in LLMs that ar… ../papers/arxiv-dcac53916c57.html https://arxiv.org/abs/2604.19502v1#2026-04-22#alignment Wed, 22 Apr 2026 11:37:03 +0800 The rapid adoption of Large Language Models (LLMs) has spurred interest in automated peer review; however, progress is currently stifled by benchmarks that treat reviewing primarily as a rating prediction task. We argue that the utility of a review lies in its textual justification--its arguments, questions, and critique--rather than a scalar score. To address this, we introduce Beyond Rating, a holistic evaluation framework that assesses AI reviewers across five dimensions: Content Faithfulnes… ../papers/arxiv-6f4a587095d1.html https://arxiv.org/abs/2604.19638v1#2026-04-22#alignment Wed, 22 Apr 2026 11:37:03 +0800 Multimodal Large Language Models are increasingly adopted as autonomous agents in interactive environments, yet their ability to proactively address safety hazards remains insufficient. We introduce SafetyALFRED, built upon the embodied agent benchmark ALFRED, augmented with six categories of real-world kitchen hazards. While existing safety evaluations focus on hazard recognition through disembodied question answering (QA) settings, we evaluate eleven state-of-the-art models from the Qwen, Gem… ../papers/arxiv-542a2e2a02e6.html https://arxiv.org/abs/2604.19405v1#2026-04-22#alignment Wed, 22 Apr 2026 11:37:03 +0800 Automatic evaluators such as reward models play a central role in the alignment and evaluation of large vision-language models (LVLMs). Despite their growing importance, these evaluators are almost exclusively assessed on English-centric benchmarks, leaving open the question of how well these evaluators generalize across languages. To answer this question, we introduce MM-JudgeBench, the first large-scale benchmark for multilingual and multimodal judge model evaluation, which includes over 60K… ../papers/arxiv-39272031a7a0.html https://arxiv.org/abs/2604.19135v1#2026-04-22#alignment Wed, 22 Apr 2026 11:37:03 +0800 This paper presents the first exploration of text-to-image diffusion models for zero-shot sketch-based 3D shape retrieval (ZS-SBSR). Existing sketch-based 3D shape retrieval methods struggle in zero-shot settings due to the absence of category supervision and the extreme sparsity of sketch inputs. Our key insight is that large-scale pretrained diffusion models inherently exhibit open-vocabulary capability and strong shape bias, making them well suited for zero-shot visual retrieval. We leverage… ../papers/arxiv-2fdaceb58972.html https://arxiv.org/abs/2604.19679v1#2026-04-22#alignment Wed, 22 Apr 2026 11:37:03 +0800 Recent advances in Diffusion Transformers (DiTs) have enabled high-quality joint audio-video generation, producing videos with synchronized audio within a single model. However, existing controllable generation frameworks are typically restricted to video-only control. This restricts comprehensive controllability and often leads to suboptimal cross-modal alignment. To bridge this gap, we present MMControl, which enables users to perform Multi-Modal Control in joint audio-video generation. MMCon… ../papers/arxiv-4d1ad4b081bb.html https://arxiv.org/abs/2604.18401v1#2026-04-21#alignment Tue, 21 Apr 2026 11:40:46 +0800 General agents have given rise to phenomenal applications such as OpenClaw and Claude Code. As these agent systems (a.k.a. Harnesses) strive for bolder goals, they demand increasingly stronger agentic capabilities from foundation Large Language Models (LLMs). Agentic Reinforcement Learning (RL) is emerging as a central post-training paradigm for empowering LLMs with these capabilities and is playing an increasingly pivotal role in agent training. Unlike single-turn token-level alignment or reas… ../papers/arxiv-5f77807f2720.html https://arxiv.org/abs/2604.18375v1#2026-04-21#alignment Tue, 21 Apr 2026 11:40:46 +0800 Conversational agents, such as ChatGPT and Doubao, have become essential daily assistants for billions of users. To further enhance engagement, these systems are evolving from passive responders to proactive companions. However, existing efforts focus on activation within ongoing dialogues, while overlooking a key real-world bottleneck. In the conversation initiation stage, users may have a vague need but no explicit query intent, creating a first-message barrier where the conversation holds be… ../papers/arxiv-ccd0dd55c2f1.html https://arxiv.org/abs/2604.17797v1#2026-04-21#alignment Tue, 21 Apr 2026 11:40:46 +0800 Referring video object segmentation (RVOS) aims to segment the target instance in a video, referred by a text expression. Conventional approaches are mostly supervised learning, requiring expensive pixel-level mask annotations. To tackle it, weakly-supervised RVOS has recently been proposed to replace mask annotations with bounding boxes or points, which are however still costly and labor-intensive. In this paper, we design a novel weakly-supervised RVOS method, namely WSRVOS, to train the mode… ../papers/arxiv-28c2e2bc1523.html https://arxiv.org/abs/2604.18562v1#2026-04-21#alignment Tue, 21 Apr 2026 11:40:46 +0800 Reasoning segmentation requires models to ground complex, implicit textual queries into precise pixel-level masks. Existing approaches rely on a single segmentation token $\texttt{<SEG>}$, whose hidden state implicitly encodes both semantic reasoning and spatial localization, limiting the model's ability to explicitly disentangle what to segment from where to segment. We introduce AnchorSeg, which reformulates reasoning segmentation as a structured conditional generation process over image toke… ../papers/arxiv-c7fa0d917c8c.html https://arxiv.org/abs/2604.18429v1#2026-04-21#alignment Tue, 21 Apr 2026 11:40:46 +0800 Change visual question answering (Change VQA) addresses the problem of answering natural-language questions about semantic changes between bi-temporal remote sensing (RS) images. Although vision-language models (VLMs) have recently been studied for temporal RS image understanding, Change VQA remains underexplored in the context of modern multimodal models. In this letter, we revisit the CDVQA benchmark using recent Qwen models under a unified low-rank adaptation (LoRA) setting. We compare Qwen3… ../papers/arxiv-1f905a979d75.html https://arxiv.org/abs/2604.18326v1#2026-04-21#alignment Tue, 21 Apr 2026 11:40:46 +0800 Recent advancements in audio-video joint generation models have demonstrated impressive capabilities in content creation. However, generating high-fidelity human-centric videos in complex, real-world physical scenes remains a significant challenge. We identify that the root cause lies in the structural deficiencies of existing datasets across three dimensions: limited global scene and camera diversity, sparse interaction modeling (both person-person and person-object), and insufficient individu… ../papers/arxiv-d700189c3374.html https://arxiv.org/abs/2604.18313v1#2026-04-21#alignment Tue, 21 Apr 2026 11:40:46 +0800 Open-Vocabulary Temporal Action Detection (OV-TAD) aims to localize and classify action segments of unseen categories in untrimmed videos, where effective alignment between action semantics and video representations is critical for accurate detection. However, existing methods struggle to mitigate the semantic imbalance between concise, abstract action labels and rich, complex video contents, inevitably introducing semantic noise and misleading cross-modal alignment. To address this challenge,… ../papers/doi-00657ec6b105.html https://pubmed.ncbi.nlm.nih.gov/42001305/#2026-04-20#alignment Mon, 20 Apr 2026 11:48:52 +0800 INTRODUCTION: Military-Civilian Partnerships (MCP) were developed to mitigate degradation of combat medical readiness during peacetime. Although these programs have historically focused on sustaining surgical readiness and training military physicians, MCP increasingly augment training for Army Combat Medics, Navy Hospital Corpsmen, Air Force Aerospace Service Specialist, and other non-physician military medical personnel. The effectiveness, scalability, and alignment of MCP along with evolving… ../papers/doi-04f076dfee40.html https://pubmed.ncbi.nlm.nih.gov/41994492/#2026-04-18#alignment Sat, 18 Apr 2026 11:26:55 +0800 PURPOSE: Vision-language models (VLMs) are increasingly used to interpret multimodal educational materials, yet their reliability on diagram-, equation-, and text-dense scientific lecture slides remains poorly understood. This work introduces Medical Imaging Lecture Understanding (MILU), a large-scale benchmark designed to characterize cross-model variability in structured understanding of real medical imaging lectures. APPROACH: MILU includes 23 lecture sets with 1117 slides. LLaVA-OneVision,… ../papers/doi-ff821e86a727.html https://pubmed.ncbi.nlm.nih.gov/41996403/#2026-04-18#alignment Sat, 18 Apr 2026 11:26:55 +0800 The rapid proliferation of Internet of Medical Things (IoMT) devices in healthcare environments has created critical cybersecurity vulnerabilities that demand both accurate and interpretable intrusion detection solutions. Existing deep learning-based intrusion detection systems (IDS) achieve high detection accuracy but lack inherent explainability, limiting their clinical adoption under regulatory frameworks such as GDPR and FDA guidelines. This paper presents MedDefender-MHAN, an explainable m… ../papers/arxiv-60286bc4afdd.html https://arxiv.org/abs/2604.15151v1#2026-04-17#alignment Fri, 17 Apr 2026 11:39:21 +0800 Large language models have demonstrated strong performance on general-purpose programming tasks, yet their ability to generate executable algorithmic trading strategies remains underexplored. Unlike standard code benchmarks, trading-strategy generation requires simultaneous mastery of domain-specific financial logic, knowledge of a specialized API, and the ability to produce code that is not only syntactically correct but also leads to actual trades on historical data. In this work, we present… ../papers/arxiv-400f736db53f.html https://arxiv.org/abs/2604.15222v1#2026-04-17#alignment Fri, 17 Apr 2026 11:39:21 +0800 Artificial Intelligence is increasingly introduced into systems engineering activities, particularly within requirements engineering, where quality assessment and validation remain heavily dependent on expert judgment. While recent AI tools demonstrate promising capabilities in analyzing and generating requirements, their role within formal systems engineering processes-and their alignment with established INCOSE criteria-remains insufficiently understood. This paper investigates the extent to… ../papers/doi-fda96f4fa371.html https://arxiv.org/abs/2604.15190v1#2026-04-17#alignment Fri, 17 Apr 2026 11:39:21 +0800 Simulating group-level user behavior enables scalable counterfactual evaluation of merchant strategies without costly online experiments. However, building a trustworthy simulator faces two structural challenges. First, information incompleteness causes reasoning-based simulators to over-rationalize when unobserved factors such as offline context and implicit habits are missing. Second, mechanism duality requires capturing both interpretable preferences and implicit statistical regularities, wh… ../papers/arxiv-4a4068542625.html https://arxiv.org/abs/2604.14951v1#2026-04-17#alignment Fri, 17 Apr 2026 11:39:21 +0800 Tool learning with foundation models aims to endow AI systems with the ability to invoke external resources -- such as APIs, computational utilities, and specialized models -- to solve complex tasks beyond the reach of standalone language generation. While recent advances in Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) have expanded their reasoning and perception capabilities, existing tool-use methods are predominantly limited to text-only inputs and closed-world s… ../papers/arxiv-0f86f7993414.html https://arxiv.org/abs/2604.14805v1#2026-04-17#alignment Fri, 17 Apr 2026 11:39:21 +0800 Grain-edge segmentation (GES) and lithology semantic segmentation (LSS) are two pivotal tasks for quantifying rock fabric and composition. However, these two tasks are often treated separately, and the segmentation quality is implausible albeit expensive, time-consuming, and expert-annotated datasets have been used. Recently, foundation models, especially the Segment Anything Model (SAM), have demonstrated impressive robustness for boundary alignment. However, directly adapting SAM to joint GES… ../papers/arxiv-283874153373.html https://arxiv.org/abs/2604.13888v1#2026-04-16#alignment Thu, 16 Apr 2026 11:43:00 +0800 The integration of Large Language Models (LLMs) into Geographic Information Systems (GIS) marks a paradigm shift toward autonomous spatial analysis. However, evaluating these LLM-based agents remains challenging due to the complex, multi-step nature of geospatial workflows. Existing benchmarks primarily rely on static text or code matching, neglecting dynamic runtime feedback and the multimodal nature of spatial outputs. To address this gap, we introduce GeoAgentBench (GABench), a dynamic and i… ../papers/arxiv-82411c54ef00.html https://arxiv.org/abs/2604.13804v1#2026-04-16#alignment Thu, 16 Apr 2026 11:43:00 +0800 The rapid evolution of multimodal large models has revolutionized the simulation of diverse characters in speech dialogue systems, enabling a novel interactive paradigm. Character attributes are manifested not only in textual responses but also through vocal features, as speech conveys rich paralinguistic information that is challenging to quantify. This poses significant difficulties in evaluating the character alignment of role-playing agents. To address these challenges, we present RoleJudge… ../papers/arxiv-7021296a66d5.html https://arxiv.org/abs/2604.13828v1#2026-04-16#alignment Thu, 16 Apr 2026 11:43:00 +0800 User simulators are essential for the scalable training and evaluation of interactive AI systems. However, existing approaches often rely on shallow user profiling, struggle to maintain persona consistency over long interactions, and are largely limited to English or single-domain settings. We present MUSE, a multi-domain Chinese user simulation framework designed to generate human-like, controllable, and behaviorally consistent responses. First, we propose Iterative Profile Self-Evolution (IPS… ../papers/arxiv-b488936a3be9.html https://arxiv.org/abs/2604.14016v1#2026-04-16#alignment Thu, 16 Apr 2026 11:43:00 +0800 Multimodal Continual Instruction Tuning (MCIT) is essential for sequential task adaptation of Multimodal Large Language Models (MLLMs) but is severely restricted by catastrophic forgetting. While existing literature focuses on the reasoning language backbone, in this work, we expose a critical yet neglected dual-forgetting phenomenon across both perception drift in Cross-modal Projection Space and reasoning collapse in Low-rank Parameter Space. To resolve this, we present \textbf{MAny} (\textbf… ../papers/arxiv-bb26f390e899.html https://arxiv.org/abs/2604.12776v1#2026-04-15#alignment Wed, 15 Apr 2026 11:35:50 +0800 Realizing endogenous narrative evolution in LLM-based multi-agent systems is hindered by the inherent stochasticity of generative emergence. In particular, long-horizon simulations suffer from social memory stacking, where conflicting relational states accumulate without resolution, and narrative-spatial dissonance, where spatial logic detaches from the evolving plot. To bridge this gap, we propose EvoSpark, a framework specifically designed to sustain logically coherent long-horizon narratives… ../papers/arxiv-df5064c793f1.html https://arxiv.org/abs/2604.12833v1#2026-04-15#alignment Wed, 15 Apr 2026 11:35:50 +0800 Vision-Language Models (VLMs) have shown remarkable performance, yet their security remains insufficiently understood. Existing adversarial studies focus almost exclusively on the digital setting, leaving physical-world threats largely unexplored. As VLMs are increasingly deployed in real environments, this gap becomes critical, since adversarial perturbations must be physically realizable. Despite this practical relevance, physical attacks against VLMs have not been systematically studied. Suc… ../papers/doi-fb5d26b2eb57.html https://pubmed.ncbi.nlm.nih.gov/41979660/#2026-04-15#alignment Wed, 15 Apr 2026 11:35:50 +0800 The use of multimodal data is essential for the precise diagnosis and treatment of brain tumors. In this context, multimodal data encompass multisequence magnetic resonance imaging, computed tomography, positron emission tomography, histopathological images, molecular and genomic profiles, structured clinical variables, and radiological reports. With the rapid advancement of artificial intelligence, integrating these heterogeneous data sources has become a central research direction for improvi… ../papers/doi-48f3f7f35ec5.html https://pubmed.ncbi.nlm.nih.gov/41979955/#2026-04-15#alignment Wed, 15 Apr 2026 11:35:50 +0800 Vision-language models in healthcare face a critical limitation, i.e., the modality gap, where image and text embeddings occupy distantly separated regions in shared representation space. This is reinforced by traditional contrastive learning objectives, and manifests itself through fundamental constraints in cross-modal understanding and downstream task performance. Existing approaches focus on addressing input-level requirements, however, the geometric constraints imposed by multimodal contra… ../papers/arxiv-a470213a0b35.html https://arxiv.org/abs/2604.11655v1#2026-04-14#alignment Tue, 14 Apr 2026 11:37:06 +0800 The rapid adoption of Large Language Models (LLMs) in interactive systems has enabled the creation of dynamic, open-ended Role-Playing Agents (RPAs). However, evaluating these agents remains a significant challenge, as standard NLP metrics fail to capture the nuances of role adherence, logical consistency, and long-term narrative stability. This paper introduces RPA-Check, a multi-stage automated evaluation framework designed to objectively assess the performance of LLM-based RPAs in complex, c… ../papers/arxiv-5cf42310e590.html https://arxiv.org/abs/2604.11806v1#2026-04-14#alignment Tue, 14 Apr 2026 11:37:06 +0800 To identify safety violations, auditors often search over large sets of agent traces. This search is difficult because failures are often rare, complex, and sometimes even adversarially hidden and only detectable when multiple traces are analyzed together. These challenges arise in diverse settings such as misuse campaigns, covert sabotage, reward hacking, and prompt injection. Existing approaches struggle here for several reasons. Per-trace judges miss failures that only become visible across… ../papers/arxiv-c894eb6a7f68.html https://arxiv.org/abs/2604.11790v1#2026-04-14#alignment Tue, 14 Apr 2026 11:37:06 +0800 Tool-augmented Large Language Model (LLM) agents have demonstrated impressive capabilities in automating complex, multi-step real-world tasks, yet remain vulnerable to indirect prompt injection. Adversaries exploit this weakness by embedding malicious instructions within tool-returned content, which agents directly incorporate into their conversation history as trusted observations. This vulnerability manifests across three primary attack channels: web and local content injection, MCP server in… ../papers/arxiv-bb16b976d1a8.html https://arxiv.org/abs/2604.11490v1#2026-04-14#alignment Tue, 14 Apr 2026 11:37:06 +0800 While the field of vision-language (VL) has achieved remarkable success in integrating visual and textual information across multiple languages and domains, there is still no dedicated framework for assessing human-centric alignment in vision-language systems. We offer two contributions to address this gap. First, we introduce Anthropogenic Regional Adaptation: a novel paradigm that aims to optimize model relevance to specific regional contexts while ensuring the retention of global generalizat… ../papers/arxiv-dd7f6721d11f.html https://arxiv.org/abs/2604.11798v1#2026-04-14#alignment Tue, 14 Apr 2026 11:37:06 +0800 Accurate delineation of the Clinical Target Volume (CTV) is essential for radiotherapy planning, yet remains time-consuming and difficult to assess, especially for complex treatments such as Total Marrow and Lymph Node Irradiation (TMLI). While deep learning-based auto-segmentation can reduce workload, safe clinical deployment requires reliable cues indicating where models may be wrong. In this work, we propose a budget-aware uncertainty-driven quality assurance (QA) framework built on nnU-Net,… ../papers/arxiv-06948b88ac9a.html https://arxiv.org/abs/2604.11788v1#2026-04-14#alignment Tue, 14 Apr 2026 11:37:06 +0800 High dynamic range (HDR) imagery offers a rich and faithful representation of scene radiance, but remains challenging for generative models due to its mismatch with the bounded, perceptually compressed data on which these models are trained. A natural solution is to learn new representations for HDR, which introduces additional complexity and data requirements. In this work, we show that HDR generation can be achieved in a much simpler way by leveraging the strong visual priors already captured… ../papers/arxiv-f426fe16d894.html https://arxiv.org/abs/2604.11775v1#2026-04-14#alignment Tue, 14 Apr 2026 11:37:06 +0800 Perturbation-based explainability methods such as KernelSHAP provide model-agnostic attributions but are typically impractical for patch-based 3D medical image segmentation due to the large number of coalition evaluations and the high cost of sliding-window inference. We present an efficient KernelSHAP framework for volumetric CT segmentation that restricts computation to a user-defined region of interest and its receptive-field support, and accelerates inference via patch logit caching, reusin… ../papers/arxiv-ba1c1d650495.html https://arxiv.org/abs/2604.06167v1#2026-04-08#alignment Wed, 08 Apr 2026 17:10:24 +0800 Churn flow-the chaotic, oscillatory regime in vertical two-phase flow-has lacked a quantitative mathematical definition for over $40$ years. We introduce the first topology-based characterization using Euler Characteristic Surfaces (ECS). We formulate unsupervised regime discovery as Multiple Kernel Learning (MKL), blending two complementary ECS-derived kernels-temporal alignment ($L^1$ distance on the $χ(s,t)$ surface) and amplitude statistics (scale-wise mean, standard deviation, max, min)-wi… ../papers/arxiv-da7fd45377bf.html https://arxiv.org/abs/2604.06156v1#2026-04-08#alignment Wed, 08 Apr 2026 17:10:24 +0800 MLLMs have been successfully applied to multimodal embedding tasks, yet their generative reasoning capabilities remain underutilized. Directly incorporating chain-of-thought reasoning into embedding learning introduces two fundamental challenges. First, structural misalignment between instance-level reasoning and pairwise contrastive supervision may lead to shortcut behavior, where the model merely learns the superficial format of reasoning. Second, reasoning is not universally beneficial for e… ../papers/arxiv-af36603bd683.html https://arxiv.org/abs/2604.06155v1#2026-04-08#alignment Wed, 08 Apr 2026 17:10:24 +0800 Whether Large Language Models (LLMs) develop coherent internal world models remains a core debate. While conventional Next-Token Prediction (NTP) focuses on one-step-ahead supervision, Multi-Token Prediction (MTP) has shown promise in learning more structured representations. In this work, we provide a theoretical perspective analyzing the gradient inductive bias of MTP, supported by empirical evidence, showing that MTP promotes the convergence toward internal belief states by inducing represen… ../papers/arxiv-29285da86277.html https://arxiv.org/abs/2604.06148v1#2026-04-08#alignment Wed, 08 Apr 2026 17:10:24 +0800 The governance of artificial intelligence has a blind spot: the machine identities that AI systems use to act. AI agents, service accounts, API tokens, and automated workflows now outnumber human identities in enterprise environments by ratios exceeding 80 to 1, yet no integrated framework exists to govern them. A single ungoverned automated agent produced $5.4-10 billion in losses in the 2024 CrowdStrike outage; nation-state actors including Silk Typhoon and Salt Typhoon have operationalized u… ../papers/arxiv-f5eea61bb28e.html https://arxiv.org/abs/2604.06124v1#2026-04-08#alignment Wed, 08 Apr 2026 17:10:24 +0800 This study proposes a lightweight multimodal adaptation framework to bridge the representation gap between RGB-pretrained VLMs and thermal infrared imagery, and demonstrates its practical utility using a real drone-collected dataset. A thermal dataset was developed from drone-collected imagery and was used to fine-tune VLMs through multimodal projector alignment, enabling the transfer of information from RGB-based visual representations to thermal radiometric inputs. Three representative models… ../papers/arxiv-da7fd45377bf.html https://arxiv.org/abs/2604.06156v1#2026-04-08#alignment Wed, 08 Apr 2026 17:10:24 +0800 MLLMs have been successfully applied to multimodal embedding tasks, yet their generative reasoning capabilities remain underutilized. Directly incorporating chain-of-thought reasoning into embedding learning introduces two fundamental challenges. First, structural misalignment between instance-level reasoning and pairwise contrastive supervision may lead to shortcut behavior, where the model merely learns the superficial format of reasoning. Second, reasoning is not universally beneficial for e… ../papers/arxiv-f5eea61bb28e.html https://arxiv.org/abs/2604.06124v1#2026-04-08#alignment Wed, 08 Apr 2026 17:10:24 +0800 This study proposes a lightweight multimodal adaptation framework to bridge the representation gap between RGB-pretrained VLMs and thermal infrared imagery, and demonstrates its practical utility using a real drone-collected dataset. A thermal dataset was developed from drone-collected imagery and was used to fine-tune VLMs through multimodal projector alignment, enabling the transfer of information from RGB-based visual representations to thermal radiometric inputs. Three representative models… ../papers/arxiv-1f39f1fd7a01.html https://arxiv.org/abs/2604.06079v1#2026-04-08#alignment Wed, 08 Apr 2026 17:10:24 +0800 Graphics Program Synthesis is pivotal for interpreting and editing visual data, effectively facilitating the reverse-engineering of static visuals into editable TikZ code. While TikZ is the de facto standard for scientific schematics due to its programmatic flexibility, its requirement for rigorous spatial precision presents a significant challenge for Multimodal Large Language Models. Progress is currently stifled by two primary gaps: (1) Data Quality Gap: existing image-TikZ corpora often lac…