Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.
arXiv:2609.08399v1 Announce Type: new Abstract: Universal machine-learning interatomic potentials (u-MLIPs) aim to generalize across diverse configurations. Benchmarks enable reproducible evaluation...
arXiv:2609.06059v1 Announce Type: new Abstract: As large language models evolve from question-answering systems into general-purpose agents, evaluation must move beyond static answer correctness to a...
arXiv:2609.09528v1 Announce Type: new Abstract: Video large language models (Video-LLMs) are increasingly used as the perceptual front end of world models, a role that assumes they can read motion: h...
arXiv:2609.06188v1 Announce Type: new Abstract: Multimodal sentiment analysis and emotion recognition in conversations demand effective modeling of heterogeneous interactions across textual, acoustic...
arXiv:2609.10317v1 Announce Type: new Abstract: Streaming talking-head generation produces each frame as its driving audio arrives, yet fidelity and efficiency have so far pulled in opposite directio...
arXiv:2609.10498v1 Announce Type: new Abstract: Recovering 3D human pose from monocular sports broadcasts remains challenging when players must be localized in a shared metric world coordinate system...
arXiv:2512.04305v3 Announce Type: replace Abstract: Vision-language models (VLMs) such as CLIP are increasingly adapted across decentralized data silos, yet the reliability of their predictions under...
arXiv:2601.07268v2 Announce Type: replace Abstract: Data-driven landslide susceptibility mapping (LSM) typically relies on landslide conditioning factors (LCFs), whose availability, heterogeneity, an...
arXiv:2605.27136v2 Announce Type: replace Abstract: Uncertainty quantification (UQ) remains a critical challenge in Large Vision Language Models (LVLMs) for reliable predictions and real-world deploy...
arXiv:2608.21244v2 Announce Type: replace Abstract: Anomaly detection aims to identify observations that deviate from normal patterns. Recent work uses pretrained vision-language models (VLMs) for tr...
arXiv:2609.09945v1 Announce Type: cross Abstract: Machine learning-based credit scoring is increasingly central to Peer-to-Peer (P2P) lending, yet its resilience to adversarial manipulation, where ap...
ComplicitSplat is a novel black‑box attack that leverages 3D Gaussian Splatting (3DGS) shading to create viewpoint‑specific camouflage, embedding adversarial content into scene objects that is only visible from certain angles. The method does not require access to model architecture or weights and can successfully fool a range of popular object detectors—including single‑stage, multi‑stage, and transformer‑based models—on both real‑world physical objects and synthetic scenes. This demonstrates that downstream models using 3DGS are vulnerable to adversarial manipulation.
FreqBLiMP is a frequency‑controlled extension of the BLiMP minimal‑pair benchmark that regenerates all 67 paradigms under explicit Zipf‑frequency regimes while preserving grammatical contrasts. The study evaluates multiple open‑weight LLM families and finds that lower lexical frequency consistently reduces sentence likelihood, yet overall contrastive acceptability accuracy drops only modestly. However, the stability in aggregate accuracy hides significant variability across linguistic phenomena, with models remaining robust on overt morphosyntactic generalization but degrading on lemma‑specific tasks.
The paper introduces Causal Visual Recurrent Reasoning (CVRR), a method that forces multimodal models to rely on latent visual states by using recurrent computation as the sole image‑conditioned path to prediction. CVRR initializes recurrence from the question hidden state after a pretrained vision‑language model has processed the image, repeatedly updates this state while re‑reading the same visual evidence, and removes all other visual traces before decoding. Experiments on several benchmarks show that CVRR maintains strong performance while other latent reasoners lose visual competence, and causal interventions confirm that predictions depend on the recurrent visual trajectory.
The paper introduces Vision-of-Thought (VoT), a framework that inserts a discrete visual-thinking layer between vision‑language models (VLMs) and diffusion transformers (DiTs). Instead of using VLMs solely as text encoders, they act as multimodal planners that generate VoT tokens—high‑level visual plans such as objects and layouts—before pixel rendering. A specialized VoT tokenizer is trained with a closed‑loop objective combining VLM alignment, feature reconstruction, and vector‑quantization losses, ensuring the tokens are semantically readable by the VLM while preserving necessary visual information. Experimental results show that VoT improves semantic alignment and offers a structured, interpretable interface for controllable generation.
A*-Thought-V2 is a framework that models Chain-of-Thought reasoning as a geometric trajectory in a 3D PCA space, using explicit-implicit latent tokens to compress steps that deviate from the main question-to-solution direction. The method measures alignment angles to decide which steps remain text and which become latent, and introduces stepwise embedding forcing and label forcing to train the architecture. Experiments on Qwen models show up to 2.6% accuracy gains, halved response length, and significant reductions in computation and training time.
SAFIRE is a large-scale benchmark for fire and smoke understanding in multimodal large language models (MLLMs), featuring 83,000 captioned images across 20 scenarios and 193,000 multiple-choice VQA questions derived from a 9.7K-image subset. The benchmark evaluates 10 dimensions of performance, from basic perception to higher-order reasoning, and employs a GPT‑5.4-assisted verification pipeline to ensure annotation quality. Experiments on ten open-source MLLMs (8B–38B) reveal an average accuracy of 61.9%, highlighting significant gaps in safety-critical reasoning, while fine-tuning vision encoders on just 7% of SAFIRE data boosts fire-scene classification accuracy from 20.1% to 64.5%. All resources are publicly available at https://risys-lab.github.io/SAFIRE/.
The IGT system tackles PolyFiQA Task 2 of the FinMMEval Lab, a multilingual financial QA challenge involving English SEC filings and news in five languages. It distinguishes two question families: numeric‑structured queries are answered via keyword extraction from filings, while synthesis queries use rule‑based passage selection from news. The approach yields a development ROUGE‑1 of ~0.395, a 60% boost over a generic RAG baseline, and places third among twelve teams on the official test set.
NormViz introduces a new benchmark, NormViz‑Bench, comprising 3,268 contrastive image pairs from 16 countries that test AI’s ability to recognize culturally relevant visual norms. Each pair differs only in a behavior that changes its cultural interpretation, and images are labeled as conforming, violating, or irrelevant to local norms, requiring both images to be correctly classified. The benchmark shows current VLMs perform poorly, and a complementary training set, NormViz‑Train, offers a path to improve performance by teaching models to link visual perception with cultural significance.
The paper introduces AeroBelief, a dual‑layer semantic‑spatial belief mapping framework for aerial object goal navigation. It separates broad contextual plausibility (intuition layer) from target‑specific evidence (evidence layer) and fuses them into persistent spatial belief hotspots. The method also employs object‑conditioned visual reasoning and egocentric regional guidance, achieving state‑of‑the‑art success rates on the UAV‑ON benchmark.