LoRA, PEFT, instruction tuning and domain adaptation — adapting a pretrained model without paying to train one.
arXiv:2607. 28959v1 Announce Type: cross Abstract: Adversarial training is one of the most effective defenses against adversarial attacks, yet the computational cost remains prohibitive at modern scales, especially for large language models (LLMs).
arXiv:2603. 25464v2 Announce Type: replace-cross Abstract: Zero-shot reinforcement learning (RL) algorithms aim to learn a family of policies from a reward-free dataset, and recover optimal policies for any reward function directly at test time.
arXiv:2607. 28639v1 Announce Type: cross Abstract: We show that knowledge distillation in small instruction-tuned language models has asymmetric effects on bias.
arXiv:2607. 29100v1 Announce Type: new Abstract: Differentially private (DP) training of text-conditioned generative models suffers a utility cliff at strong privacy.
arXiv:2607. 29287v1 Announce Type: cross Abstract: Multi-domain machine translation (MDMT) poses a unique challenge due to varying levels of linguistic complexity across domains.
arXiv:2603. 17205v3 Announce Type: replace-cross Abstract: Domain-specific finetuning is essential for dense retrievers, yet not all data pairs contribute equally to the learning process.
arXiv:2607. 28658v1 Announce Type: cross Abstract: Federated pre-training offers a way to train foundation models on private or distributed data without centralizing the underlying datasets.
arXiv:2607. 28680v1 Announce Type: cross Abstract: Entity linking in tables matches short and ambiguous cell mentions to their corresponding knowledge-base entities.
Despite the progress of multimodal large language models (MLLMs), they continue to exhibit deficiencies in visual perception. Following visual instruction tuning, internal MLLM representations rapidly deviate from their original semantic states during inference, causing severe information degradation.
Modern sensing, communication, and learning systems generate heterogeneous network signals, with local data differing in dimension, modality, and geometric structure. Processing such data requires a mathematical framework capable of simultaneously modeling heterogeneous local signal spaces and the transformations relating them.
Query-agnostic KV cache eviction compresses a context once and reuses the resulting cache for arbitrary future queries, but performance can collapse under tight budgets. Existing methods primarily improve which original KV pairs are retained.
Deploying Vision-Language-Action (VLA) robots as mobile edge nodes within wireless sensor networks (WSNs) requires robust protection against physical adversarial threats. We present VLAGuard, a framework to assess and mitigate a critical vulnerability: policy-critical action-to-vision attention hijacking.
With growing privacy and portability concerns, source-free domain adaptation requires only a source pre-trained model and an unlabeled target domain, allowing for effective adaptation to the target data. Most existing self-training methods focus on selecting and exploiting samples with reliable predictions, often neglecting others.
arXiv:2607. 26432v1 Announce Type: cross Abstract: Face anti-spoofing (FAS) is increasingly expected to provide not only bona fide/spoof decisions, but also attack semantics and image-grounded evidence for human inspection.
arXiv:2607. 26389v1 Announce Type: cross Abstract: Fine-tuning a language model on data containing a narrow flaw, such as insecure code or incorrect mathematical answers, can cause broad misalignment through a mechanism that remains debated.
arXiv:2607. 26368v1 Announce Type: cross Abstract: Financial disclosures contain numerical claims, temporal statements, entity references, policy commitments, and risk descriptions that may conflict in qualitatively different ways.
arXiv:2607. 27539v1 Announce Type: new Abstract: Exact deletion from persistent language-model memory depends on how that memory represents a record.
arXiv:2607. 27594v1 Announce Type: new Abstract: Post-training alignment in large reasoning models (LRMs) has significantly improved their adaptability to diverse safety compliance settings.
arXiv:2607. 26119v1 Announce Type: new Abstract: Large reasoning models trained via reinforcement learning (RL) have been increasingly shown to outperform their supervised fine-tuned (SFT) counterparts on mathematical reasoning tasks; Yet the mechanistic basis for this advantage remains unclear.
arXiv:2607. 27271v1 Announce Type: new Abstract: Code models are increasingly trained with execution feedback, but most training signals still stop at correctness.