Efficient red-teaming method to uncover vulnerabilities in Large Language Models (LLMs) is crucial. While recent attacks often use LLMs as optimizers, the discrete language space make gradient-based methods struggle. We introduce LARGO (Latent Adversarial Reflection through Gradient Optimization), a novel latent self-reflection attack that reasserts the power of gradient-based optimization for generating fluent jailbreaking prompts. By operating within the LLM's continuous latent space, LARGO first optimizes an adversarial latent vector and then recursively call the same LLM to decode the latent into natural language. This methodology yields a fast, effective, and transferable attack that produces fluent and stealthy prompts.

Transformer architectures have achieved remarkable success across language, vision, and multimodal tasks, and there is growing demand for them to address in-context compositional learning tasks. In these tasks, models solve the target problems by inferring compositional rules from context examples, which are composed of basic components structured by underlying rules. However, some of these tasks remain challenging for Transformers, which are not inherently designed to handle compositional tasks and offer limited structural inductive bias. In this work, inspired by the principle of sparse coding, we propose a reformulation of the attention to enhance its capability for compositional tasks. In sparse coding, data are represented as sparse combinations of dictionary atoms with coefficients that capture their compositional rules.

Despite the impressive generative abilities of black-box large language models (LLMs), their inherent opacity hinders further advancements in capabilities such as reasoning, planning, and personalization. Existing works aim to enhance LLM capabilities via domain-specific adaptation, which require additional training on accessible model parameters, an infeasible option for black-box LLMs. To address this challenge, we introduce Matryoshka Pilot(M-Pilot), a lightweight white-box LLM controller that guides a large-scale black-box LLM generator by decomposing complex tasks into a series of intermediate outputs. Specifically, we consider the black-box LLM as an environment, with M-Pilot serving as a policy to provide intermediate guidance through prompts for driving the black-box LLM. M-Pilot is trained to pivot the outputs of the black-box LLM aligning with preferences during iterative interaction, which enables controllable multi-turn generation and self-improvement in optimizing intermediate guidance. Empirical evaluations on diverse tasks demonstrate that our method effectively enhances the capabilities of black-box LLMs in complex, long-horizon tasks.

The ability to selectively remove knowledge from LLMs is highly desirable. However, existing methods often struggle with balancing unlearning efficacy and retain model utility, and lack controllability at inference time to emulate base model behavior as if it had never seen the unlearned data. In this paper, we propose LUNAR, a novel unlearning method grounded in the Linear Representation Hypothesis and operates by redirecting the representations of unlearned data to activation regions that expresses its inability to answer. We show that contrastive features are not a prerequisite for effective activation redirection, and LUNARachieves state-of-the-art unlearning performance and superior controllability. Specifically, LUNARachieves between 2.9 and 11.7 improvement in the combined unlearning efficacy and model utility score (Deviation Score) across various base models and generates coherent, contextually appropriate responses post-unlearning. Moreover, LUNAR effectively reduces parameter updates to a single down-projection matrix, a novel design that significantly enhances efficiency by 20 and robustness. Finally, we demonstrate that LUNARis robust to white-box adversarial attacks and versatile in real-world scenarios, including handling sequential unlearning requests.

Egocentric video reasoning centers on an unobservable agent behind the camera who dynamically shapes the environment, requiring inference of hidden intentions and recognition of fine-grained interactions. This core challenge limits current multimodal large language models (MLLMs), which excel at visible event reasoning but lack embodied, first-person understanding. To bridge this gap, we introduce EgoThinker, a novel framework that endows MLLMs with robust egocentric reasoning capabilities through spatio-temporal chain-ofthought supervision and a two-stage learning curriculum. First, we introduce EgoRe-5M, a large-scale egocentric QA dataset constructed from 13M diverse egocentric video clips. This dataset features multi-minute segments annotated with detailed CoT rationales and dense hand-object grounding. Second, we employ SFT on EgoRe-5M to instill reasoning skills, followed by reinforcement fine-tuning (RFT) to further enhance spatio-temporal localization. Experimental results show that EgoThinker outperforms existing methods across multiple egocentric benchmarks, while achieving substantial improvements in finegrained spatio-temporal localization tasks.

As large language models (LLMs) are increasingly deployed in high-stakes settings, the risk of generating harmful or toxic content remains a central challenge. Post-hoc alignment methods are brittle: once unsafe patterns are learned during pretraining, they are hard to remove. In this work, we present a data-centric pretraining framework that builds safety into the model from the start. Our framework consists of four key steps: (i) Safety Filtering: building a safety classifier to classify webdata into safe and unsafe categories; (ii) Safety Rephrasing: we recontextualize unsafe webdata into safer narratives; (iii) Native Refusal: we synthetically generate pretraining datasets that actively teach models to refuse on unsafe content and the moral reasoning behind it, and (iv) Harmfulness-Tag annotated pretraining: we flag unsafe content during pretraining using a special token, and use it to steer models away from unsafe generations at inference-time. Our safety-pretrained models reduce attack success rates from 38.8% to 8.4% on standard LLM safety benchmarks with no performance degradation on general tasks.

Gradient-based data attribution methods, such as influence functions, are critical for understanding the impact of individual training samples without requiring repeated model retraining. However, their scalability is often limited by the high computational and memory costs associated with per-sample gradient computation. In this work, we propose GRASS, a novel gradient compression algorithm and its variants FACTGRASS for linear layers specifically, that explicitly leverage the inherent sparsity of per-sample gradients to achieve sub-linear space and time complexity. Extensive experiments demonstrate the effectiveness of our approach, achieving substantial speedups while preserving data influence fidelity. In particular, FACTGRASS achieves up to 165% faster throughput on billion-scale models compared to the previous state-of-the-art baselines.

For models trained for multi-image input, text prompt is:850 Which objects are present in both images? You can think of your answer in any way (e.g. For models where we first concatenate the input images, the text prompt is:855 There are two images provided, one on the left and the other on the right.856 Which objects are present in both images? You can think of your answer in any way (e.g. We used the following procedure to guide our creation of images.

Multimodal language models possess a remarkable ability to handle an openvocabulary worth of objects. Yet the best models still suffer from hallucinations when reasoning about scenes in the real world, revealing a gap between their seemingly strong performance on existing perception benchmarks that are saturating and their reasoning in the real world. To address this gap, we build a novel benchmark of in-the-wild scenes that we call Common-OBench. With more than 10.5k examples using exclusively new images not found in web training data to avoid contamination, Common-OBenchgoes beyond just perception, inspired by cognitive tests for humans, to probe reasoning across scenes by asking "what's in common?". We evaluate leading multimodal language models, including models specifically trained to reason. We find that perceiving objects in single images is easy for most models, yet reasoning across scenes is very challenging even for the best models, including reasoning models. Despite saturating many leaderboards focusing on perception, the best performing model only achieves 35% on Common-OBench--and on Common-OComplex, consisting of more complex scenes, the best model achieves only 1%. Curiously, we find models are more prone to hallucinate when similar objects are present in the scene, suggesting models may be relying on object co-occurrence seen during training. Among the models we evaluated, we found scale can provide modest improvements while models explicitly trained with multi-image inputs show bigger improvements, suggesting scaled multi-image training may offer promise.

Low-Rank Adaptation (LoRA) has become the de facto parameter-efficient finetuning (PEFT) method for large language models (LLMs) by constraining weight updates to low-rank matrices. Recent works such as Tied-LoRA, VeRA, and VBLoRA push efficiency further by introducing additional constraints to reduce the trainable parameter space. In this paper, we show that the parameter space reduction strategies employed by these LoRA variants can be formulated within a unified framework, Uni-LoRA, where the LoRA parameter space, flattened as a highdimensional vector space RD, can be reconstructed through a projection from a subspace Rd, with d D. We demonstrate that the fundamental difference among various LoRA methods lies in the choice of the projection matrix, P RD d. Most existing LoRA variants rely on layer-wise or structure-specific projections that limit cross-layer parameter sharing, thereby compromising parameter efficiency. In light of this, we introduce an efficient and theoretically grounded projection matrix that is isometric, enabling global parameter sharing and reducing computation overhead. Furthermore, under the unified view of Uni-LoRA, this design requires only a single trainable vector to reconstruct LoRA parameters for the entire LLM - making UniLoRA both a unified framework and a "one-vector-only" solution. Extensive experiments on GLUE, mathematical reasoning, and instruction tuning benchmarks demonstrate that Uni-LoRA achieves state-of-the-art parameter efficiency while outperforming or matching prior approaches in predictive performance.