相关论文: Exploiting Positional Bias for Query-Agnostic Gene…

Language models are instruction-tuned to refuse harmful requests, but the mechanisms underlying this behavior remain poorly understood. Popular steering methods operate on the residual stream and degrade output coherence at high…

Most existing unbiased learning-to-rank (ULTR) approaches are based on the user examination hypothesis, which assumes that users will click a result only if it is both relevant and observed (typically modeled by position). However, in…

This paper concerns corpus poisoning attacks in dense information retrieval, where an adversary attempts to compromise the ranking performance of a search algorithm by injecting a small number of maliciously generated documents into the…

What happens when a language model thinks without words? Standard reasoning LLMs verbalize intermediate steps as chain-of-thought; latent reasoning transformers (LRTs) instead perform deliberation entirely in continuous hidden space. We…

Large Language Models (LLMs) often struggle to maintain their original performance when faced with semantically coherent but task-irrelevant contextual information. Although prior studies have explored this issue using fixed-template or…

Recent work has demonstrated the vulnerability of modern text classifiers to universal adversarial attacks, which are input-agnostic sequences of words added to text processed by classifiers. Despite being successful, the word sequences…

Ranking relevance is a fundamental task in search engines, aiming to identify the items most relevant to a given user query. Traditional relevance models typically produce scalar scores or directly predict relevance labels, limiting both…

Personalized search provides a potentially powerful tool, however, it is limited due to the large number of roles that a person has: parent, employee, consumer, etc. We present the role-relevance algorithm: a search technique that favors…

Although pre-trained language models (PrLMs) have achieved significant success, recent studies demonstrate that PrLMs are vulnerable to adversarial attacks. By generating adversarial examples with slight perturbations on different levels…

The context window of large language models has been extended to 128k tokens or more. However, language models still suffer from position bias and have difficulty in accessing and using the middle part of the context due to the lack of…

Large Language Models (LLMs) increasingly rely on automatic prompt engineering in graphical user interfaces (GUIs) to refine user inputs and enhance response accuracy. However, the diversity of user requirements often leads to unintended…

Textual adversarial attacks mislead NLP models, including Large Language Models (LLMs), by subtly modifying text. While effective, existing attacks often require knowledge of the victim model, extensive queries, or access to training data,…

Presentation bias is one of the key challenges when learning from implicit feedback in search engines, as it confounds the relevance signal with uninformative signals due to position in the ranking, saliency, and other presentation factors.…

Safety alignment in large language models (LLMs) induces over-refusals -- where LLMs decline benign requests due to aggressive safety filters. We analyze this phenomenon in retrieval-augmented generation (RAG), where both the query intent…

Dense retrieval is a crucial task in Information Retrieval (IR), serving as the basis for downstream tasks such as re-ranking and augmenting generation. Recently, large language models (LLMs) have demonstrated impressive semantic…

Dense retrieval systems have been widely used in various NLP applications. However, their vulnerabilities to potential attacks have been underexplored. This paper investigates a novel attack scenario where the attackers aim to mislead the…

Web-use agents are rapidly being deployed to automate complex web tasks with extensive browser capabilities. However, these capabilities create a critical and previously unexplored attack surface. This paper demonstrates how attackers can…

Training robust deep learning models for down-stream tasks is a critical challenge. Research has shown that down-stream models can be easily fooled with adversarial inputs that look like the training data, but slightly perturbed, in a way…

Retrieval-augmented generation and tool-integrated LLM agents increasingly depend on external textual sources. This reliance broadens the available attack surface, allowing adversaries to insert malicious instructions that trigger…

We investigate whether LLMs display a well-known human cognitive bias, the attraction effect, in hiring decisions. The attraction effect occurs when the presence of an inferior candidate makes a superior candidate more appealing, increasing…