The Power of Noise: Redefining Retrieval for RAG Systems

Retrieval-Augmented Generation (RAG) systems represent a significant advancement over traditional Large Language Models (LLMs). RAG systems enhance their generation ability by incorporating external data retrieved through an Information Retrieval (IR) phase, overcoming the limitations of standard LLMs, which are restricted to their pre-trained knowledge and limited context window. Most research in this area has predominantly concentrated on the generative aspect of LLMs within RAG systems. Our study fills this gap by thoroughly and critically analyzing the influence of IR components on RAG systems. This paper analyzes which characteristics a retriever should possess for an effective RAG's prompt formulation, focusing on the type of documents that should be retrieved. We evaluate various elements, such as the relevance of the documents to the prompt, their position, and the number included in the context. Our findings reveal, among other insights, that including irrelevant documents can unexpectedly enhance performance by more than 30% in accuracy, contradicting our initial assumption of diminished quality. These results underscore the need for developing specialized strategies to integrate retrieval with language generation models, thereby laying the groundwork for future research in this field.

Adversarial Data Poisoning for Fake News Detection: How to Make a Model Misclassify a Target News without Modifying It

Fake news detection models are critical to countering disinformation but can be manipulated through adversarial attacks. In this position paper, we analyze how an attacker can compromise the performance of an online learning detector on specific news content without being able to manipulate the original target news. In some contexts, such as social networks, where the attacker cannot exert complete control over all the information, this scenario can indeed be quite plausible. Therefore, we show how an attacker could potentially introduce poisoning data into the training data to manipulate the behavior of an online learning method. Our initial findings reveal varying susceptibility of logistic regression models based on complexity and attack type.

A graph neural network-based model with Out-of-Distribution Robustness for enhancing Antiretroviral Therapy Outcome Prediction for HIV-1

Predicting the outcome of antiretroviral therapies for HIV-1 is a pressing clinical challenge, especially when the treatment regimen includes drugs for which limited effectiveness data is available. This scarcity of data can arise either due to the introduction of a new drug to the market or due to limited use in clinical settings. To tackle this issue, we introduce a novel joint fusion model, which combines features from a Fully Connected (FC) Neural Network and a Graph Neural Network (GNN). The FC network employs tabular data with a feature vector made up of viral mutations identified in the most recent genotypic resistance test, along with the drugs used in therapy. Conversely, the GNN leverages knowledge derived from Stanford drug-resistance mutation tables, which serve as benchmark references for deducing in-vivo treatment efficacy based on the viral genetic sequence, to build informative graphs. We evaluated these models' robustness against Out-of-Distribution drugs in the test set, with a specific focus on the GNN's role in handling such scenarios. Our comprehensive analysis demonstrates that the proposed model consistently outperforms the FC model, especially when considering Out-of-Distribution drugs. These results underscore the advantage of integrating Stanford scores in the model, thereby enhancing its generalizability and robustness, but also extending its utility in real-world applications with limited data availability. This research highlights the potential of our approach to inform antiretroviral therapy outcome prediction and contribute to more informed clinical decisions.

RRAML: Reinforced Retrieval Augmented Machine Learning

The emergence of large language models (LLMs) has revolutionized machine learning and related fields, showcasing remarkable abilities in comprehending, generating, and manipulating human language. However, their conventional usage through API-based text prompt submissions imposes certain limitations in terms of context constraints and external source availability. To address these challenges, we propose a novel framework called Reinforced Retrieval Augmented Machine Learning (RRAML). RRAML integrates the reasoning capabilities of LLMs with supporting information retrieved by a purpose-built retriever from a vast user-provided database. By leveraging recent advancements in reinforcement learning, our method effectively addresses several critical challenges. Firstly, it circumvents the need for accessing LLM gradients. Secondly, our method alleviates the burden of retraining LLMs for specific tasks, as it is often impractical or impossible due to restricted access to the model and the computational intensity involved. Additionally we seamlessly link the retriever's task with the reasoner, mitigating hallucinations and reducing irrelevant, and potentially damaging retrieved documents. We believe that the research agenda outlined in this paper has the potential to profoundly impact the field of AI, democratizing access to and utilization of LLMs for a wide range of entities.

Investigating the Robustness of Sequential Recommender Systems Against Training Data Perturbations: an Empirical Study

Sequential Recommender Systems (SRSs) have been widely used to model user behavior over time, but their robustness in the face of perturbations to training data is a critical issue. In this paper, we conduct an empirical study to investigate the effects of removing items at different positions within a temporally ordered sequence. We evaluate two different SRS models on multiple datasets, measuring their performance using Normalized Discounted Cumulative Gain (NDCG) and Rank Sensitivity List metrics. Our results demonstrate that removing items at the end of the sequence significantly impacts performance, with NDCG decreasing up to 60\%, while removing items from the beginning or middle has no significant effect. These findings highlight the importance of considering the position of the perturbed items in the training data and shall inform the design of more robust SRSs.

Integrating Item Relevance in Training Loss for Sequential Recommender Systems

Sequential Recommender Systems (SRSs) are a popular type of recommender system that learns from a user's history to predict the next item they are likely to interact with. However, user interactions can be affected by noise stemming from account sharing, inconsistent preferences, or accidental clicks. To address this issue, we (i) propose a new evaluation protocol that takes multiple future items into account and (ii) introduce a novel relevance-aware loss function to train a SRS with multiple future items to make it more robust to noise. Our relevance-aware models obtain an improvement of ~1.2% of NDCG@10 and 0.88% in the traditional evaluation protocol, while in the new evaluation protocol, the improvement is ~1.63% of NDCG@10 and ~1.5% of HR w.r.t the best performing models.

Encoding Concepts in Graph Neural Networks

The opaque reasoning of Graph Neural Networks induces a lack of human trust. Existing graph network explainers attempt to address this issue by providing post-hoc explanations, however, they fail to make the model itself more interpretable. To fill this gap, we introduce the Concept Encoder Module, the first differentiable concept-discovery approach for graph networks. The proposed approach makes graph networks explainable by design by first discovering graph concepts and then using these to solve the task. Our results demonstrate that this approach allows graph networks to: (i) attain model accuracy comparable with their equivalent vanilla versions, (ii) discover meaningful concepts that achieve high concept completeness and purity scores, (iii) provide high-quality concept-based logic explanations for their prediction, and (iv) support effective interventions at test time: these can increase human trust as well as significantly improve model performance.