The Latest Papers about AI
-
DTCA: Denoising Token Credit Assignment for Reinforcement Learning in Diffusion Language Models
2026-10-26 10:54:11Abstract
Diffusion language models (DLMs) generate text through iterative denoising, enabling bidirectional context modeling and parallel token refinement. Recent studies have applied reinforcement learning (RL) to DLMs and improved their reasoning performance. However, the iterative denoising process makes credit assignment more challenging, as final outcome rewards provide limited information about how different tokens contribute to the final output. In this paper, we propose Denoising Token Credit Assignment (DTCA), a framework for token-level credit assignment in reinforcement learning for diffusion language models. DTCA estimates token contributions by measuring the improvement in prediction confidence of remaining masked positions after token commitment and groups these signals according to part-of-speech (POS) categories. These contribution weights are used to select informative denoising timesteps and assign token-level learning signals during RL training. DTCA further calibrates these learning signals according to token-level entropy changes induced by policy updates. Experiments on LLaDA-8B-Instruct show that DTCA consistently improves mathematical reasoning performance across different generation budgets, with gains of up to 9.0 points on MATH500 and 5.3 points on GSM8K over the base model.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35201
PDF
-
Trustworthy Recursive Self-Improvement in the Era of Large Models: Foundation, Risks, Evaluation and Open Challenges
Abstract
AI systems built around large models increasingly use experience and feedback to make persistent changes to models, prompts, memories, tools, and workflows. Recursive self-improvement (RSI) extends this process to inductive biases and to the mechanisms that govern future updates, which can themselves be revised and reused across improvement cycles. Yet self-modification alone does not guarantee trustworthy improvement. Reported gains may reflect flawed evaluations, previously acquired capabilities may deteriorate, and compliance with behavioral constraints may weaken. Compromised artifacts or sensitive information may also persist in successor systems. Changes to improvement mechanisms may invalidate earlier verification results and make human oversight less effective. This survey presents an operational definition and process framework for RSI and organizes the requirements for trustworthy RSI into four interconnected themes: genuine and sustained improvement; behavioral constraints, alignment, and fairness; system security and privacy; and transparency and controllability. We examine these themes from three perspectives: Improvement Validity, Constraint Preservation, and Effective Oversight. This analysis covers improvement diagnosis, proposal generation, proposal evaluation, and update implementation (S1--S4), including the recursive application of these stages to improvement mechanisms (RI). We review risks, safeguards, and evaluation evidence for individual updates, effects inherited by successor systems, and revisions to improvement mechanisms, while distinguishing direct RSI findings from evidence in related systems. Finally, we examine evaluation settings, identify requirements for independent and longitudinal assessment, and discuss applications and open research problems.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35193
PDF
-
Lost in Translation: Investigating Systematic Discrepancies between Parallel English and Chinese Names of American Chinese Restaurants
Abstract
Given that the Chinese are one of the oldest ethnic groups to immigrate to the US, their cuisine have become central to the US dining scene and an interesting avenue to study how Chinese immigrants have culturally adapted into American life. In that regard, previous research has focused on analyzing the English names that Chinese restaurants adopt, e.g., finding the use of familiar and typical words. To investigate how Chinese immigrants have culturally integrated in the US depending on social, geographic and demographic contexts, we investigate in this work the under-explored relationship between the Chinese and English names of American Chinese restaurants. Specifically, we conduct a study on 3,172 Chinese restaurants (10% of the total on Google Maps) sampled across the US. We present 8-dimension name frames to capture the underlying meanings behind names and create the NameFrames dataset containing 884 annotated parallel Chinese/English names. We also built an NLP pipeline to analyze the semantic discrepancies between names with both frame-based and embedding-based distance metrics. Our frame analysis shows that English names more often emphasize Location and Specialty, while Chinese names more often emphasize Positivity and Ambiance. Furthermore, we observe that for Chinese restaurants located in urban areas, those in counties with a higher concentration of Chinese population, or those that are relatively new, the discrepancies between the English and Chinese names are noticeably larger. Finally, interviews with 10 Chinese restaurant owners reveal additional interesting insights regarding naming conventions.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35191
PDF
-
V$^3$AST-Bench: Verifiable, Valid, and Versatile Safety Evaluation for Coding Agents
Abstract
Large language models increasingly act through code, modifying repositories, executing programs, and interacting with services in agent form. Their actions introduce risks that cannot be assessed from generated responses alone. Safety evaluation must verify actual consequences, distinguish harm avoidance from task incapability, and cover realistic workflows. We introduce V$^3$AST-Bench, a benchmark for Verifiable results, Valid interpretation, and Versatile risk coverage of Agent Safety. It comprises 212 risk tasks spanning 18 categories. Dedicated runtime environments verify harm and task outcomes through executable checks over states, records, and artifact behavior, while paired capability twin tasks verify workflow competence. Our experiments span 51 combinations of 11 models and five harnesses, spanning 86,496 trials and over 9 billion tokens of model interaction. Results show that V$^3$AST-Bench remains challenging even for frontier coding agents, as Claude Opus 4.8 with Claude Code still causes verified harm on 20.9% of tasks. Across model-harness combinations, agents with comparable competence exhibit markedly different safety outcomes, showing that task capability alone does not guarantee safe execution. More importantly, even explicit recognition of risks does not reliably prevent harmful actions, revealing a critical gap between risk awareness and safe execution. These findings highlight the need to move beyond risk recognition toward enforcing safety throughout execution.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35188
PDF
-
High Scores, Hidden Failures: Evaluating Codebook-Based LLM Content Analysis
Abstract
Large language models (LLMs) are increasingly used for codebook-based content analysis in computational social science, where codebooks translate theoretical concepts into labeling schemes. While prior work reports strong predictive performance, it remains unclear whether these systems produce reliable and robust text annotations. We conduct a large-scale empirical failure-mode analysis of LLM-based coding pipelines across four datasets and five models by systematically evaluating their stability, robustness, and conceptual validity under controlled perturbations, domain shifts, and optimization procedures. We uncover multiple failure modes that are largely invisible under standard metrics, showing that high aggregate scores frequently coexist with severe label-level disparities, instability across runs, domain sensitivity, and shifts in label interpretation. Importantly, these failure modes persist under widely used optimization strategies and remain largely invisible under standard metrics such as accuracy and micro-F1. Our results demonstrate that current LLM-based coding pipelines cannot be treated as stable measurement tools without targeted validation beyond aggregate performance.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35179
PDF
-
RetroNav: Scene-Grounded Retrospection and Retry for Self-Evolving Lifelong Navigation
Abstract
Existing robot navigation methods primarily focus on isolated tasks, overlooking the need to accumulate knowledge and experience across repeated navigation episodes. Although recent lifelong navigation methods have explored cross-task knowledge accumulation, they mainly focus on scene knowledge while neglecting historical navigation experience and its connection to scene evidence, limiting the ability to reflect on past mistakes and improve future decisions. To address this limitation, this paper presents RetroNav, a training-free self-evolving navigation framework that integrates navigation, memory accumulation, and failure reflection. Specifically, we develop a unified memory system that connects geometric and semantic scene knowledge with historical navigation experience, distinguishing objective observations from task-dependent interpretations. After each navigation attempt, an independent evaluator determines whether a retry is necessary. If so, a Retrospection Agent revisits historical decisions using accumulated scene knowledge and navigation records, generating an evidence-grounded retry brief to guide the next attempt. RetroNav achieves state-of-the-art performance across five navigation benchmarks. Our method will be open-sourced at this https URL .
Abstract (translated)
URL
https://arxiv.org/abs/2610.35172
PDF
-
Formal Explanations for Linear Temporal Logic over Finite Traces
Abstract
Linear Temporal Logic over Finite Traces (LTLf) is widely used in Artificial Intelligence to specify desired temporal behavior and determine whether finite traces satisfy or violate it. Explaining these outcomes is essential for debugging and trust. However, existing approaches typically capture either sufficient reasons or counterfactual causes, without clarifying their relationship. We introduce a native characterization of abductive (AXps) and contrastive (CXps) explanations directly over finite traces, based on locking and unlocking time-indexed events. We prove its equivalence to the classifier-based formulation, and show that CXps coincide with a recently proposed temporal causality semantics for LTLf. This characterization enables an efficient implementation based on the enumeration of minimal unsatisfiable cores and minimal correction sets of ASP programs. Experiments on real-world event logs demonstrate the effectiveness of the approach.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35169
PDF
-
Do Language Models Understand Igbo Culture? Beyond Multiple Choice
Abstract
The value of language models for any community depends on whether a model actually understands the culture of the people it intends to serve. We use the Igbo language as a case study. Existing benchmarks test translated Western ideas, lacking depth. To address this, we introduce CAB-Igbo, a benchmark developed from indigenous cultural sources and verified by native speakers across six domains. We evaluated diverse models under varying conditions, incorporating regional personas. Performance was assessed using multiple-choice accuracy and human evaluations across Epistemic Fidelity, Representational Richness, Linguistic Quality, and Pragmatic Proficiency. Results reveal a stark dissociation between language fluency and cultural knowledge. Through statistical testing, we show that for most models, the severe lack of Epistemic Fidelity and Representational Richness is already present in English and does not significantly worsen when switching to Igbo. Instead, native language prompting mainly degrades Linguistic Quality and Pragmatic Proficiency. Gemini 2.0 Flash is the only model whose cultural knowledge scores improve under Igbo prompting. We find that current models struggle to demonstrate the knowledge needed for true alignment with Igbo, and multiple-choice formats allow pattern-matching to appear as understanding. Furthermore, automated judges agree with humans on linguistic quality but fail on epistemic fidelity. Ultimately, cultural evaluation must be built from community knowledge, establishing human evaluation as the gold standard to prevent misrepresenting low-resource communities globally.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35168
PDF
-
How Tuning, Reference Preview, Actuator Modeling, and Fallback Policies Change Quadrotor Controller Comparisons: A Reproducible Simulation Study
Abstract
Quadrotor controller comparisons commonly report a single tracking error, although the resulting ranking also depends on tuning effort, reference information, actuator prediction, optimizer fallback, and command timing. We quantify these effects for cascade PID, hover LQR, boundary-layer sliding-mode control (BL-SMC), linear MPC with LQR fallback (LMPC-F), a preview-enabled LMPC-F, and a geometric $SO(3)$ controller on a common nonlinear plant with rotor gyroscopic coupling, motor lag, and saturation. A primary study (556 episodes at 50~Hz) combines matched twelve-candidate tuning, 400 paired evaluation episodes, and a 48-cell MPC factorial; a confirmation study (1462 episodes) repeats the evaluation on 50 fresh realizations with tuning budgets of 12, 24, and 48 candidates. A matched twelve-candidate tuning protocol changed the mean-RMSE ordering obtained with the fixed baseline settings, moving BL-SMC from last to first; the new ordering recurred on the fresh realizations, and a 48-candidate budget left it unchanged. Reference preview substantially reduced MPC tracking error and, on the figure-eight task, removed fallback activity. On an abrupt step, the outcome depended on the interplay between fallback policy, predictor, and actuator authority. Command timing also changed the outcome: replaying measured command-call delays terminated three of four nominal MPC configurations on a desktop computer and all four on a Raspberry Pi~5, whereas capping the solver at 100 iterations removed every deadline miss in the tested nominal timing profiles. Controller rankings therefore depend not only on tracking accuracy but also on the tuning protocol, reference information, fallback behavior, and implementation timing. We propose a reporting checklist covering tracking accuracy, tuning effort, completion and failure rates, fallback activity, applied-input variation, and command timing.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35165
PDF
-
AutoEnergyDesign: An LLM-Driven Agent for Automatic Energy System Design Optimization
Abstract
Designing building energy systems is a high-dimensional, mixed-integer search problem: when each design is evaluated by a year-long simulation, black-box optimizers typically need hundreds of evaluations. This paper presents AutoEnergyDesign, a framework in which a Large Language Model (LLM) agent reasons over previous simulation results, proposes configuration changes and learns in context from each outcome, while a git-based mechanism automatically reverts unsuccessful changes. The agent couples a reasoning model with an explicit memory organisation (procedural, semantic, episodic and reflective), a constrained action interface and an environment-grounded keep/revert signal. The simulator covers eight technologies (photovoltaics, micro-wind, combined heat and power, solar thermal, gas boilers, batteries, heat pumps and inverters) and computes the 25-year Net Present Cost (NPC), including capital and operating costs and salvage value. On a case study with measured building load profiles, the agent, run with Claude Haiku 5.5, Sonnet 5.5 and Opus 5.5 (ten runs of 50 simulations each), is compared with random search, Tree-structured Parzen Estimator (TPE) Bayesian optimization and a $(1+1)$ evolution strategy sharing the agent's acceptance rule (300 evaluations, ten seeds). With all three models the agent reaches an NPC of about 3.53 MEUR, 1.20% below the best TPE result and 0.23% below the best evolution-strategy result. It comes within 1% of this design after a median of 18 (Opus) to 27.5 (Haiku) simulations, whereas among the baselines only the evolution strategy does so, in half of its seeds and after a median of 192 evaluations. Opus reaches the best design in all ten runs, Sonnet in six (two runs never come within 1%), and Haiku comes within 1% in every run but reaches it exactly only once. Every decision is recorded in an auditable trace.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35160
PDF
-
Complexity-guided Regularization for Generalizable Human Gaussian Splatting
Abstract
Recent studies on generalizable human Gaussian splatting have achieved the notable progress in synthesizing novel views of unseen human subjects from multi-view images. Despite this progress, existing methods still struggle to represent fine-grained details of human subjects, which inherently involve high geometric and textural complexity. This is because previous methods do not impose a constraint on how densely Gaussians are placed, which often leads to blurry rendering results in complex regions. To address this problem, we propose a complexity-guided regularization scheme. The key idea is to leverage a complexity prior computed via the local entropy in geometric and textural characteristics of a clothed human. This effectively prevents Gaussians from being coarsely placed in complex regions, which facilitates the faithful reconstruction of intricate structures. Furthermore, we sample local patches predominantly from complex regions according to this complexity prior, which allows the model to be sufficiently supervised by local details. Experimental results on benchmark datasets demonstrate that our method delivers state-of-the-art performance while maintaining the real-time inference speed in generalizable human Gaussian splatting. The code and model are publicly available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35152
PDF
-
SoK: Timestamps Are Not a Time Model: Rethinking Temporal Reasoning in Host-Based Intrusion Detection
Abstract
Host-based intrusion detectors order events, window audit streams and sequence attack stages by their timestamps. Every conclusion they draw therefore rests on assumptions about clocks, delivery, entity identity and the integrity of the evidence, and most of these assumptions go unstated. This Systematization of Knowledge (SoK) makes them explicit. Drawing on temporal databases, stream processing and runtime verification, we distill five components of a time model, together with a checklist of evidence assumptions and four levels of guarantee, and walk through a running campaign with them. A documented search yields fifteen families of detection and investigation. For papers selected by screening role and period, plus named systems and landmarks, we check the stated assumptions in the full texts, and we assess eleven temporal knowledge representation and reasoning (KRR) formalisms and a learned baseline on nine attack patterns. Some early alert correlators stated clock and delivery assumptions, and forensics measures clock error. Current provenance-based detectors state no clock bound. Four audited papers describe input-ordering conditions or conventions, but none specifies conditions for conclusion validity across a stated class of delivery failures. In the selected provenance-based systems, campaign composition uses system-specific procedures or objectives, learned state, causal graph queries or language models, and beyond CEP-based query systems, we found no temporal rule language or stream reasoner from KRR applied to host audit data. In 28 of 38 audited provenance-era detector papers, capture or stored evidence is trusted or its manipulation is placed out of scope, and none says which conclusions survive deleted, delayed or forged records. Five open challenges follow from these gaps.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35151
PDF
-
LongSearchEval: Evaluating Long-Horizon Web Search Capabilities of Large Language Models via a Trajectory-Aware Agent Framework with Failure-Node Localization
Abstract
Long-horizon web search requires LLM-based agents to integrate information across web pages through extended interactions under task-specific constraints. Evaluating this capability is important not only for measuring final answer quality, but also for assessing whether an agent acquires, verifies, and uses evidence reliably over extended interactions. Existing evaluations, however, mainly rely on simulated trajectories or on post-hoc trajectories reconstructed from final answers, rather than on retained real interaction traces, which limits the fidelity of behavioral assessment and weakens support for fine-grained error localization. To address this gap, we introduce LONGSEARCHEVAL, a trajectory-aware agent evaluation framework that combines a unified agent execution protocol with source constrained outcome verification and retrospective failure-node localization. Stage 1 verifies answer groups against agent-cited webpages without human-authored gold answers; Stage 2 audits normalized rounds and links flagged failure nodes to incorrect answer groups. We evaluate 16 models from eight families on 100 tasks, yielding 1,600 retained runs and 6,545 automatically flagged failure nodes. Trajectory collection and evaluation required 1,037 and 318 hours of cumulative runtime, respectively. Human validation yields 94.50% agreement on supplied answer-group labels and 98.11% feature-correspondence precision for audited round-group links. Together, outcome scores and linked trajectory evidence support controlled model comparison and retrospective inspection of answer-related content.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35150
PDF
-
Temporal Slice Learning for AI-Generated Video Detection with 400$times$ Fewer FLOPs
Abstract
Recent advances in generative video models have improved visual realism, making detection increasingly dependent on subtle temporal inconsistencies. However, these signals are weak and spatially localized, while most of the video content is redundant. This creates a fundamental mismatch: existing detectors rely on heavy backbones to model full videos, which leads to high computational cost and limited scalability. In this work, we rethink AI-generated video detection from a representation perspective. Instead of modeling entire videos, we restructure them into compact temporal slice representations that isolate informative temporal evolution. We introduce a lightweight framework using frequency-aware temporal slice consistency learning, where informative spatial rows and columns are aggregated over time to form slice-domain inputs. This reparameterization suppresses redundant appearance while preserving discriminative temporal dynamics, enabling efficient and targeted detection. By constructing slices from both high- and low-frequency regions, our method exposes complementary temporal artifacts, including unstable details and abnormal motion smoothness. We instantiate this idea in a Temporal Slice Consistency Network, which integrates task-driven slice localization, hierarchical positional encoding, and a lightweight Transformer to model cross-slice and cross-frequency dependencies. We further introduce AI-Artist, a new benchmark with artist-curated videos from recent high-fidelity generators, including Seedance and Kling. Experiments on GenVideo and AI-Artist show that our method achieves strong performance while requiring over 400$\times$ fewer model-forward FLOPs and 3.6$\times$ faster model inference than the strong video-based baseline ReStraV (2.9$\times$ faster end to end, including preprocessing), enabling scalable and practical deployment.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35148
PDF
-
Locked In: Multi-Objective RL for the AI Data Center Build-Out
Abstract
AI data center build-out is a multi-trillion-dollar, multi-decade infrastructure program. Yet siting, power procurement, and cooling decisions are still made with static, single-point scoring tools. These tools are evaluated once, on day-zero numbers. This is a poor fit. A site's water, power, and carbon profile drifts substantially over the 10 to 20 year asset life. The governing decisions are also irreversible, and they bind capital and physical plant for years. We formalize long-horizon build-out as a tiered multi-objective Markov decision process. Strategic commitments are gated to discrete windows, incur switching costs, and energize only after source-dependent lead times. Operational dispatch is restricted to capacity that physically exists. We solve the problem with a preference-conditioned multi-objective agent that has a semi-Markov manager and worker hierarchy. The agent recovers a family of policies that a planner can steer across cost, carbon, water, and reliability at inference time. We validate long-horizon dynamics with a lightweight counterfactual rollout model. Across three commitment regimes (irreversible, reversible, and commit-once) on real site data, preference conditioning yields a confirmed, steerable trade-off on both objective pairs in every regime. Imposing lock-in produces a more evenly covered frontier and a far more consistent rate of anticipatory retrofit ahead of a binding water-stress constraint than a freely reversible formulation. Finally, a compact counterfactual dynamics model tracks slow-moving state well. It is measurably outperformed by simple persistence on the water-stress axis. We report this limitation rather than obscure it. We release the environment, agents, baselines, and evaluation harness. All of them are grounded in real interconnection-queue, water-stress, and grid-emissions data.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35147
PDF
-
Sexual dimorphism and sex-specific features of pediatric hand bone shapes via statistical shape modeling
Abstract
Understanding the envelope of shape variation encompassing both sexual dimorphism and growth-related changes in living tissue is a non-trivial task. In this work, we use statistical shape models (SSMs) and analyze demographic-stratified shape parameters to quantify sexual dimorphism and sex-specific shape changes. To this end, we developed a two-stage detection-segmentation framework that accurately segmented $n = 27664$ unique bone masks from a radiographic dataset of $n = 1456$ subjects with diverse ages and sexes (HD95 = $1.512 \pm 0.005$, ASSD $0.449 \pm 001$, Dice = $0.987 \pm 0.00004$, IoU = $0.975 \pm 0.0001$). Following SSM fitting, our follow-up analyses of stratified shape parameters highlighted the morphology of sexual dimorphisms for each bone and demonstrated sex-specific differences in shape development. Our results were consistent with existing literature and provided an unambiguous, data-driven methodology for characterizing tissue shape and its population-scale variation. This approach provides a potential methodological foundation for developing population-based models of anatomical variation and could be extended to investigate tissue growth and shape changes over time. Future work should evaluate the applicability of this methodology on alternative anatomical structures and imaging modalities, with the goal of validating these growth trajectories on longitudinal data.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35143
PDF
-
CalibraRL: Confidence-Calibrated Reinforcement Learning for Mitigating Hallucinations in RAG
Abstract
Large language models (LLMs) often produce plausible but incorrect answers, motivating reinforcement learning methods that reward correct answers, penalize incorrect answers, and allow abstention. However, fixed outcome-based rewards do not distinguish incorrect responses with different confidence levels or account for differences in answering reliability across queries. Consequently, such rewards may inadequately penalize overconfident errors and fail to discourage unnecessary abstention. We propose CalibraRL, a reinforcement learning framework that incorporates response confidence and query-level answering evidence into reward design. For non-abstaining responses, CalibraRL combines correctness-based utility with a Brier-based penalty that encourages agreement between confidence and correctness. For abstentions, it estimates conditional answering reliability using a Beta posterior derived from the correctness of non-abstaining responses within each sampled group. The abstention reward increases with the posterior probability that the model's accuracy on non-abstaining responses is below a predefined threshold. CalibraRL reuses training rollouts without requiring additional annotations indicating whether the model knows the answer to each question. Experiments on three QA datasets demonstrate that CalibraRL improves answer precision and reduces hallucinations compared with strong baselines. Further analysis shows that CalibraRL reduces mean expected calibration error (ECE) by 30.8\% across two backbone models and three datasets, highlighting its potential to improve the reliability of retrieval-augmented generation (RAG) in web search.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35142
PDF
-
Co-Design Bench: Multimodal Collaborative Coding Evaluation for Agentic Models
Abstract
Evaluating human-AI co-design requires testing whether coding agents can translate real users' multimodal requests into working application revisions. We introduce Co-Design Bench, a reference-free benchmark of 273 tasks curated from 37,463 deduplicated queries across 26,314 real user coding sessions. Its two tracks, Multimodal Editing (ME) and Repository Editing (RE), cover localized edits with optional image, audio, video, or document inputs and coordinated repository revisions. Tasks span three capability dimensions: Grounded Understanding, UI Transformation, and Functional Implementation. The tasks pair normalized requests and runnable pre-edit applications with 2,331 hidden expert-authored acceptance criteria. Our judge-then-refute protocol grades criteria using source and browser evidence, challenges positive verdicts with a separate LLM verifier, and audits visual and interaction claims without gold code or target screenshots. Under a shared agentic harness, the strongest of ten model endpoints passes an average of 69.9% of criteria per task but fully satisfies only 26.4% of tasks. All ten endpoints achieve lower criterion completion on repository edits spanning three or four change types than on those spanning one or two. Co-Design Bench provides a testbed for studying the gap between partial requirement satisfaction and complete task success under agentic execution for real-user application revisions.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35138
PDF
-
Domain-Adapted Latent Diffusion Models for CT-to-PET Image Translation
Abstract
Positron Emission Tomography (PET) provides important metabolic information for tumor detection and staging but remains significantly less accessible than Computed Tomography (CT) due to its high cost, infrastructure requirements, and reliance on injected radioactive tracers. CT-to-PET image translation using generative models aims to address this disparity, yet state-of-the-art diffusion-based methods are limited by the scarcity of paired training data. To overcome this, we propose a domain-adapted latent diffusion framework that transfers natural-image priors using a relatively small dataset of 21,930 paired CT/PET slices, incorporating a signed power transform to resolve extreme intensity sparsity in PET data. Compared to CPDM, a state-ofthe-art baseline for CT-to-PET image translation, our approach yields superior performance in pixel-wise reconstruction and localized tumor fidelity, demonstrating effective transfer of natural-image priors to the medical domain. We further provide a systematic analysis of evaluation metric reliability for sparse PET data, showing that standard full-image metrics can be misleading indicators of translated image quality, and propose background-filtered and mask-restricted metrics as more clinically meaningful alternatives.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35132
PDF
-
When Fusion Order Matters: Insights on Progressive Multimodal Representation Learning for Cross-modal Retrieval
Abstract
Multimodal video retrieval combines visual, acoustic, and speech information, yet how their fusion order should depend on the data and the retrieval query remains unclear. We introduce Unified Progressive Representation Learning for Order Analysis, or UniPRO, to investigate this question through configurable fusion schedules with matched model components and training settings. Starting from text-to-video retrieval, we find that the tested one-by-one schedules outperform stacked fusion, while changing the order within one-by-one fusion also changes performance. To guide order selection, we examine two complementary considerations: learning difficulty under unequal modality availability and preservation of information relevant to the query. Complete-modality controls show that missing inputs alone cannot explain order sensitivity. At R@1, text and audio retrieval favor schedules that introduce the modality matching the query last. Controlled speech missingness during training and testing reverses the preferred order in both R@1 and SumR, with different crossover points. These findings motivate balancing the two considerations when selecting a fusion path. We then use them to predict a visual-last schedule for image queries, which achieves 97.9 R@1, the highest among the evaluated orders on MSR-VTT-9k. Main experiments on four text-to-video benchmarks establish the effectiveness of UniPRO, with component ablations assessing its supporting design. Together, the results identify fusion order as a task-dependent choice that should account jointly for modality availability and the retrieval query.
Abstract (translated)
URL
https://arxiv.org/abs/2610.35125
PDF
