LoRA, PEFT, instruction tuning and domain adaptation — adapting a pretrained model without paying to train one.
arXiv:2507.02735v4 Announce Type: replace-cross Abstract: Prompt injection attacks, where untrusted data contains an injected prompt to manipulate the system, have been listed as the top security thr...
arXiv:2609.36813v1 Announce Type: new Abstract: Large language models (LLMs) exhibit strong general capabilities that mechanistic interpretability has attributed to sparse computational circuits. How...
arXiv:2609.36766v1 Announce Type: new Abstract: Can a fixed continuous prefix replace a given low-rank adapter while the attention head stays frozen? In this research, we show that the answer depends...
Vision‑Language Models (VLMs) used in autonomous driving are evaluated on their ability to understand velocity through three tasks: surrounding‑agent speed, current ego speed, and short‑horizon future ego‑speed. Experiments on nuScenes show that while current ego speed can be accessed internally, verbal outputs are poor and surrounding‑agent speed is weakly encoded; temporal cues and frame order are largely ignored. Specialized driving models (Alpamayo‑1.5) improve latent speed representations but still leave gaps between internal knowledge and verbal readout, indicating that good planning outputs do not guarantee reliable dynamic state recovery.
arXiv:2609.36826v1 Announce Type: new Abstract: Video generation models have demonstrated emerging zero-shot capabilities for visual reasoning, perception, and other vision tasks. However, diffusion-...
arXiv:2609.36659v1 Announce Type: new Abstract: The strong generalization performance of on-policy post-training paradigms has motivated studies of their parameter update behaviors. However, these st...
arXiv:2609.36616v1 Announce Type: new Abstract: Historical street-view imagery records urban evolution, but uneven coverage leaves substantial gaps in historical records. Generating plausible past ap...
arXiv:2609.36913v1 Announce Type: cross Abstract: Transcribing domain-specific entities and rare proper nouns remains a major challenge in automatic speech recognition (ASR). In this paper, we propos...
arXiv:2609.37855v1 Announce Type: cross Abstract: Hand anthropometry supports protective-glove design, but existing measurement methods often require trained operators, specialized hardware, or manua...
arXiv:2609.36644v1 Announce Type: new Abstract: Cross-attention is a crucial component in learning-based image-to-point-cloud (I2P) registration. Although existing cross-attention mechanisms have ach...
arXiv:2606.28249v2 Announce Type: replace-cross Abstract: Recently, Large Language Model (LLM)-based Text-to-Speech (TTS) models have achieved remarkable naturalness. However, the standard Supervised...
arXiv:2608.03275v2 Announce Type: replace Abstract: Parameter-efficient fine-tuning (PEFT) enables efficient adaptation of large language models, but existing MoE-based PEFT methods typically improve...
arXiv:2609.36219v1 Announce Type: new Abstract: Perspective taking is a fundamental component of spatial intelligence, requiring models interpret spatial relations from a specified viewpoint, such as...
Beacon is a new agentic visual reasoning model that improves multimodal large language models (MLLMs) by better deciding when to use tools and how to use them. It introduces two key concepts—Mode Adaptiveness, which ensures tools are invoked only when necessary, and Tool Effect, which measures the net benefit of tool use— and trains the model with supervised fine‑tuning and reinforcement learning that rewards necessity-aware decisions and expands capability through expert hints. Across 13 benchmarks, Beacon outperforms other open‑source models, achieving the highest average score and the largest net tool‑gain on diagnostic tests.
arXiv:2609.36563v1 Announce Type: new Abstract: Visual emotion recognition commonly assumes that all evidence required for prediction is contained in the observed image or video. Yet the same visible...
arXiv:2609.11149v3 Announce Type: replace-cross Abstract: How fast does a language model degrade when trained on its own outputs? Theory traces it to gradually accumulating errors, while experiments...
Approximate machine unlearning seeks to remove the influence of a forget set from a trained model without full retraining. Existing gradient-based methods require data-dependent hyperparameter search,...
Direct-decision models turn text into low-latency structured labels and scores, making them attractive for classification and automatic evaluation. Yet reliability requires more than accuracy: a model...
Computer use agents (CUAs) have demonstrated strong capabilities in completing digital tasks. However, existing CUAs either rely solely on graphical user interface (GUI) interactions, which are often...
Multimodal continual instruction tuning (MCIT) aims to enable multimodal large language models to acquire new capabilities from sequential tasks while preserving previously learned knowledge. Existing...