Multimodal instruction tuning has proven to be an effective strategy for achieving zero-shot generalization by fine-tuning pre-trained Large Multimodal Models (LMMs) with instruction-following data. However, as the scale of LMMs continues to grow, fully fine-tuning these models has become highly parameter-intensive. Although Parameter-Efficient Fine-Tuning (PEFT) methods have been introduced to reduce the number of tunable parameters, a significant performance gap remains compared to full fine-tuning. Furthermore, existing PEFT approaches are often highly parameterized, making them difficult to interpret and control. In light of this, we introduce Multimodal Representation Tuning (MRT), a novel approach that focuses on directly editing semantically rich multimodal representations to achieve strong performance and provide intuitive control over LMMs. Empirical results show that our method surpasses current state-of-the-art baselines with significant performance gains (e.g., 1580.40 MME score) while requiring substantially fewer tunable parameters (e.g., 0.03% parameters). Additionally, we conduct experiments on editing instrumental tokens within multimodal representations, demonstrating that direct manipulation of these representations enables simple yet effective control over network behavior.

Despite their remarkable potential, Large Vision-Language Models (LVLMs) still face challenges with object hallucination, a problem where their generated outputs mistakenly incorporate objects that do not actually exist. Although most works focus on addressing this issue within the language-model backbone, our work shifts the focus to the image input source, investigating how specific image tokens contribute to hallucinations. Our analysis reveals a striking finding: a small subset of image tokens with high attention scores are the primary drivers of object hallucination. By removing these hallucinatory image tokens (only 1.5% of all image tokens), the issue can be effectively mitigated. This finding holds consistently across different models and datasets. Building on this insight, we introduce EAZY, a novel, training-free method that automatically identifies and Eliminates hAllucinations by Zeroing out hallucinatorY image tokens. We utilize EAZY for unsupervised object hallucination detection, achieving 15% improvement compared to previous methods. Additionally, EAZY demonstrates remarkable effectiveness in mitigating hallucinations while preserving model utility and seamlessly adapting to various LVLM architectures.

Pre-trained language models (PLMs) have achieved impressive results on various natural language processing tasks. However, recent research has revealed that these models often rely on superficial features and shortcuts instead of developing a genuine understanding of language, especially for natural language understanding (NLU) tasks. Consequently, the models struggle to generalize to out-of-domain data. In this work, we propose Divergence Based Regularization (DBR) to mitigate this shortcut learning behavior. Our method measures the divergence between the output distributions for original examples and examples where shortcut tokens have been masked. This process prevents the model's predictions from being overly influenced by shortcut features or biases. We evaluate our model on three NLU tasks and find that it improves out-of-domain performance with little loss of in-domain accuracy. Our results demonstrate that reducing the reliance on shortcuts and superficial features can enhance the generalization ability of large pre-trained language models.

Generative AI models, renowned for their ability to synthesize high-quality content, have sparked growing concerns over the improper generation of copyright-protected material. While recent studies have proposed various approaches to address copyright issues, the capability of large vision-language models (LVLMs) to detect copyright infringements remains largely unexplored. In this work, we focus on evaluating the copyright detection abilities of state-of-the-art LVLMs using a various set of image samples. Recognizing the absence of a comprehensive dataset that includes both IP-infringement samples and ambiguous non-infringement negative samples, we construct a benchmark dataset comprising positive samples that violate the copyright protection of well-known IP figures, as well as negative samples that resemble these figures but do not raise copyright concerns. This dataset is created using advanced prompt engineering techniques. We then evaluate leading LVLMs using our benchmark dataset. Our experimental results reveal that LVLMs are prone to overfitting, leading to the misclassification of some negative samples as IP-infringement cases. In the final section, we analyze these failure cases and propose potential solutions to mitigate the overfitting problem.

Large language models (LLMs) have achieved remarkable success in contextual knowledge understanding. In this paper, we show that these concentrated massive values consistently emerge in specific regions of attention queries (Q) and keys (K) while not having such patterns in values (V) in various modern transformer-based LLMs (Q, K, and V mean the representations output by the query, key, and value layers respectively). Through extensive experiments, we further demonstrate that these massive values play a critical role in interpreting contextual knowledge (i.e., knowledge obtained from the current context window) rather than in retrieving parametric knowledge stored within the model's parameters. Our further investigation of quantization strategies reveals that ignoring these massive values leads to a pronounced drop in performance on tasks requiring rich contextual understanding, aligning with our analysis. Finally, we trace the emergence of concentrated massive values and find that such concentration is caused by Rotary Positional Encoding (RoPE), which has appeared since the first layers. These findings shed new light on how Q and K operate in LLMs and offer practical insights for model design and optimization.

Rare diseases are challenging to diagnose due to limited patient data and genetic diversity. Despite advances in variant prioritization, many cases remain undiagnosed. While large language models (LLMs) have performed well in medical exams, their effectiveness in diagnosing rare genetic diseases has not been assessed. To identify causal genes, we benchmarked various LLMs for gene prioritization. Using multi-agent and Human Phenotype Ontology (HPO) classification, we categorized patients based on phenotypes and solvability levels. As gene set size increased, LLM performance deteriorated, so we used a divide-and-conquer strategy to break the task into smaller subsets. At baseline, GPT-4 outperformed other LLMs, achieving near 30% accuracy in ranking causal genes correctly. The multi-agent and HPO approaches helped distinguish confidently solved cases from challenging ones, highlighting the importance of known gene-phenotype associations and phenotype specificity. We found that cases with specific phenotypes or clear associations were more accurately solved. However, we observed biases toward well-studied genes and input order sensitivity, which hindered gene prioritization. Our divide-and-conquer strategy improved accuracy by overcoming these biases. By utilizing HPO classification, novel multi-agent techniques, and our LLM strategy, we improved causal gene identification accuracy compared to our baseline evaluation. This approach streamlines rare disease diagnosis, facilitates reanalysis of unsolved cases, and accelerates gene discovery, supporting the development of targeted diagnostics and therapies.

Understanding how large language models (LLMs) acquire, retain, and apply knowledge remains an open challenge. This paper introduces a novel framework, K-(CSA)^2, which categorizes LLM knowledge along two dimensions: correctness and confidence. The framework defines six categories of knowledge, ranging from highly confident correctness to confidently held misconceptions, enabling a nuanced evaluation of model comprehension beyond binary accuracy. Using this framework, we demonstrate how techniques like chain-of-thought prompting and reinforcement learning with human feedback fundamentally alter the knowledge structures of internal (pre-trained) and external (context-dependent) knowledge in LLMs. CoT particularly enhances base model performance and shows synergistic benefits when applied to aligned LLMs. Moreover, our layer-wise analysis reveals that higher layers in LLMs encode more high-confidence knowledge, while low-confidence knowledge tends to emerge in middle-to-lower layers.

Large Language Models (LLMs) are known to be vulnerable to backdoor attacks, where triggers embedded in poisoned samples can maliciously alter LLMs' behaviors. In this paper, we move beyond attacking LLMs and instead examine backdoor attacks through the novel lens of natural language explanations. Specifically, we leverage LLMs' generative capabilities to produce human-readable explanations for their decisions, enabling direct comparisons between explanations for clean and poisoned samples. Our results show that backdoored models produce coherent explanations for clean inputs but diverse and logically flawed explanations for poisoned data, a pattern consistent across classification and generation tasks for different backdoor attacks. Further analysis reveals key insights into the explanation generation process. At the token level, explanation tokens associated with poisoned samples only appear in the final few transformer layers. At the sentence level, attention dynamics indicate that poisoned inputs shift attention away from the original input context during explanation generation. These findings enhance our understanding of backdoor mechanisms in LLMs and present a promising framework for detecting vulnerabilities through explainability.

Recent advancements in Large Language Models (LLMs) have showcased their impressive inference capabilities in handling complex natural language tasks that require both extensive knowledge and sophisticated reasoning abilities (OpenAI, 2024; Touvron et al., 2023; Wei et al., 2022a). LLMs have demonstrated the ability to memorize vast amounts of knowledge, and techniques like Chain-of-Thought (CoT) (Wei et al., 2022b), Tree of thoughts (ToT) (Yao et al., 2024) have been developed to further enhance their inference abilities by decomposing complex problems into several simpler, single-step processes. These methods enable LLMs to tackle multi-step inference tasks more effectively by organizing the thought process into discrete, focused actions (Feng et al., 2024; Jin et al., 2024b; Wei et al., 2022b). However, despite these advancements, existing inference frameworks often operate as an opaque process without explicit separation between knowledge retrieval and reasoning steps. This makes it unclear what specific knowledge the model utilizes and how it performs reasoning, leaving the decision-making process ambiguous. For complex, knowledge-intensive tasks, such as multi-hop inference, LLMs often struggle to effectively leverage their memory for inference (Yang et al., 2023; Jin et al., 2024b; Wang et al., 2024b; Cheng et al., 2024; Liu et al., 2024). Such tasks typically require the ability to recall relevant knowledge for each reasoning step (or "hop") and then perform inference over that recalled memory (Wang et al., 2024c). The lack of structure in the output and effective memory utilization can lead to issues such as hallucinations, where LLMs generate plausible but incorrect information (Xu et al., 2024; Li et al., 2024a), and "forgetting," where relevant information is lost across reasoning steps (Jin et al., 2024b; Chen & Shu, 2023), disrupting the logical flow.

Recently in robotics, Vision-Language-Action (VLA) models have emerged as a transformative approach, enabling robots to execute complex tasks by integrating visual and linguistic inputs within an end-to-end learning framework. While VLA models offer significant capabilities, they also introduce new attack surfaces, making them vulnerable to adversarial attacks. With these vulnerabilities largely unexplored, this paper systematically quantifies the robustness of VLA-based robotic systems. Recognizing the unique demands of robotic execution, our attack objectives target the inherent spatial and functional characteristics of robotic systems. In particular, we introduce an untargeted position-aware attack objective that leverages spatial foundations to destabilize robotic actions, and a targeted attack objective that manipulates the robotic trajectory. Additionally, we design an adversarial patch generation approach that places a small, colorful patch within the camera's view, effectively executing the attack in both digital and physical environments. Our evaluation reveals a marked degradation in task success rates, with up to a 100\% reduction across a suite of simulated robotic tasks, highlighting critical security gaps in current VLA architectures. By unveiling these vulnerabilities and proposing actionable evaluation metrics, this work advances both the understanding and enhancement of safety for VLA-based robotic systems, underscoring the necessity for developing robust defense strategies prior to physical-world deployments.