Skip to main content
Search

Nankai University

11 related articles

Subscribe to updates
  1. arXiv

    Remote-sensing fusion stops optimizing pixels alone: Lance closes a text loop and lifts detection mAP by 6.4 on a held-out detector

    Remote-sensing image fusion and semantic interpretation have usually been optimized separately, so good fusion metrics need not preserve downstream retrieval or detection utility. Lance maps each input to a closed-vocabulary text record of scene, objects and layout, conditions fusion on that language, re-describes the fused image as text, and back-propagates through text consistency and an anti-collapse text–mask contrastive loss with negative queries. Across seven test sets on GF2, QuickBird, WorldView-III and WorldView-II, reduced-scale reference metrics improve by up to 11.7% (median 2.1%) over the strongest learning baseline, and a held-out RemoteCLIP verifier plus a supervised DOTA detector give +6.4 mAP@0.5 over the closest competitor.

  2. arXiv

    Streaming video agents stop recording only what is present: DeltaStreamer logs frame-to-frame changes and scores 66.07% on OVO-Bench

    Streaming video understanding must retain evidence before anyone knows which details later queries will need, yet existing methods flatten event transitions when compressing history and cannot recover details once a frame has passed. DeltaStreamer explicitly records visual changes between adjacent frames in a hierarchical textual memory, and after a query arrives a query-driven planner decides whether to retrieve, keep observing, or replay historical frames. It scores 66.07% overall on OVO-Bench and 59.53% on StreamingBench, above the representative streaming methods it is compared with.

  3. arXiv

    Similar agent skills stop being judged by description alone: a capability graph scores the weakest link and lifts ALFWorld success from 22% to 29%

    As skill libraries grow, semantically similar skills become hard to tell apart from descriptions alone. SKALOS builds a multi-hierarchy capability graph of capability, method, artifact and constraint nodes, uses Max–Min witness learning to require that one implementation jointly covers every query requirement, and calibrates extraction errors with ranking feedback. On SkillRet its NDCG@10 is 2.68 points above a matched text-only reranker; on controlled ALFWorld tasks, witness-guided skill selection raises success from 22% to 29%.

  4. arXiv

    Ad creatives stop being redrawn whole: editing only the attributes that drag clicks lifts CTR 6.86% over 20 million impressions

    E-commerce ad image optimization moves from regenerating the whole creative to fine-grained attribute editing: 28 click-sensitive visual attributes are selected, editing suggestions are aligned to real click preferences, and a one-week online test over roughly 20 million impressions lifts aggregate CTR by 6.86%.

  5. arXiv

    Position Forcing recovers token positions from VecSet latents and quantizes them progressively, reaching best or tied-best ULIP and Uni3D scores for single-stage 3D generation

    Observing that VecSet latent tokens retain recoverable spatial correspondences, the work proposes Position Forcing: during denoising it recovers token positions from the predicted clean latent, quantizes them at progressively finer resolutions according to the noise level, and feeds them back to the diffusion Transformer through 3D RoPE, providing coarse-to-fine spatial guidance for single-stage 3D generation without a separate position generation stage and achieving best or tied-best results on all four ULIP and Uni3D metrics.

  6. arXiv

    Ego2World compiles egocentric cooking videos into executable planning environments, where six planners on 105 tasks often had accepted operations leave goals unmet

    The authors introduce Ego2World, a benchmark that compiles annotated cooking activities into executable planning environments under partial observation, linking source steps and objects to symbolic action rules, persistent world states, and explicit task conditions while maintaining world state and agent belief separately; evaluating six planners on 105 tasks shows accepted operations often leave task goals unmet, and execution traces and condition checks distinguish interrupted runs, partial attainment, and completed execution without goal attainment; in a separate paired Qwen-Plus study, persistent belief improves action validity by 4.15 percentage points and reduces visual-query attempts by 90.27%, with higher token use and no detected completion gain.

  7. arXiv

    SynLat aligns compression boundaries to syntax and retains 94.0% of uncompressed accuracy at roughly 44% compression

    SynLat reframes chain-of-thought compression as a representation-fidelity allocation problem: constituency syntax partitions a trace into ordered, non-overlapping Syntax-Aligned Units (SAUs), an answer-conditioned Teacher builds progressive KEEP/LATENT targets for a single compression-conditioned Student, and latent units map to a shared discrete codebook, so at inference the Student generates mixed text-and-latent reasoning from only the question and requested compression level. Across two Qwen3 Student scales, Standard- and Long-CoT groups, and three compression levels, SynLat matches or exceeds the strongest evaluated baseline in all 12 task-group aggregates and strictly leads in 11, with overall gains of 3.6/2.6 points at MEDIUM and 7.0/5.5 points at HIGH.

  8. Nankai University team's 362-day Internet-wide scan finds exposed Ollama endpoints grew 53.3% in a year while only 0.43%–2.90% of below-fix IPs upgraded in place

    Using XMap to probe port 11434 across the full IPv4 space daily from 2025-02-14 to 2026-02-13 over 362 observed days, a Nankai University team found 152,137 cumulative exposed Ollama endpoints, with daily alive endpoints rising from 10,473 to 16,059 (+53.3%), 26.4% of IPs appearing on a single day, only 0.43%–2.90% of below-fix IPs upgrading in place across five CVE cutoffs, the top five countries/regions holding about 70% of weighted observations, and the top five ASNs all cloud or hosting providers, complemented by PTR and port-443 TLS probing of operational characteristics.

  9. arXiv

    Pruned CTC removes linear-in-vocabulary memory growth from large-vocabulary CTC training, and LLM-CTC stays within 7% relative WER of LLM-CE on GigaSpeech while recognizing 7–10× faster

    The work introduces Pruned CTC, which exploits the fact that every valid CTC alignment uses only target tokens and blank to restrict alignment computation to that batch-level subset while retaining full-vocabulary normalization, and proves this vocabulary reduction is exactly equivalent to full-vocabulary CTC in loss and first-order gradients; with a Zipformer-M encoder and a 180K vocabulary it cuts full-step memory by 5.1× at only 17% step-time overhead and matches standard CTC accuracy across three corpora; built on it, LLM-CTC adapts pretrained causal LLMs to non-autoregressive ASR over native vocabularies, staying within 7% relative WER of LLM-CE across six Qwen3 sizes from 0.

  10. Rational design of advanced electrocatalysts based on reactivity descriptors for high-performance lithium-sulfur batteries

    This review organizes lithium-sulfur battery catalyst research around the fundamental chemistry of sulfur conversion and its rate-limiting steps, links catalyst modulation strategies to descriptors, groups descriptors into electronic, thermodynamic, and structural categories with their property-performance relationships, and points toward universal descriptors as well as the potential roles of in situ characterization, computational modeling, and artificial intelligence in descriptor construction and the design of highly active catalysts.