Bridging the Preference Gap between Retrievers and LLMs

Large Language Models (LLMs) have demonstrated superior results across a wide range of tasks, while retrieval has long been established as an effective means of obtaining task-relevant information for humans. Retrieval-augmented Generation (RAG) are known for their effectiveness in knowledge-intensive tasks by locating relevant information and placing it within the context window of the LLM. However, the relationship between retrievers and LLMs is still under-investigated. Most existing work treats the retriever and the LLM as independent components and leaves a gap between retrieving human-friendly information and assembling a LLM-friendly context. In this work, we examine a novel bridge model, validate the ranking and selection assumptions in retrievers in the context of RAG, and propose a training framework that chains together supervised and reinforcement learning to learn a bridge model. Empirical results demonstrate the effectiveness of our method in both question-answering and personalized generation tasks.

Vertical Federated Image Segmentation

With the popularization of AI solutions for image based problems, there has been a growing concern for both data privacy and acquisition. In a large number of cases, information is located on separate data silos and it can be difficult for a developer to consolidate all of it in a fashion that is appropriate for machine learning model development. Alongside this, a portion of these localized data regions may not have access to a labelled ground truth. This indicates that they have the capacity to reach conclusions numerically, but are not able to assign classifications amid a lack of pertinent information. Such a determination is often negligible, especially when attempting to develop image based solutions that often necessitate this capability. With this being the case, we propose an innovative vertical federated learning (VFL) model architecture that can operate under this common set of conditions. This is the first (and currently the only) implementation of a system that can work under the constraints of a VFL environment and perform image segmentation while maintaining nominal accuracies. We achieved this by utilizing an FCN that boasts the ability to operate on federates that lack labelled data and privately share the respective weights with a central server, that of which hosts the necessary features for classification. Tests were conducted on the CamVid dataset in order to determine the impact of heavy feature compression required for the transfer of information between federates, as well as to reach nominal conclusions about the overall performance metrics when working under such constraints.

Flexibly Scaling Large Language Models Contexts Through Extensible Tokenization

Large language models (LLMs) are in need of sufficient contexts to handle many critical applications, such as retrieval augmented generation and few-shot learning. However, due to the constrained window size, the LLMs can only access to the information within a limited context. Although the size of context window can be extended by fine-tuning, it will result in a substantial cost in both training and inference stage. In this paper, we present Extensible Tokenization as an alternative method which realizes the flexible scaling of LLMs' context. Extensible Tokenization stands as a midware in between of the tokenized context and the LLM, which transforms the raw token embeddings into the extensible embeddings. Such embeddings provide a more compact representation for the long context, on top of which the LLM is able to perceive more information with the same context window. Extensible Tokenization is also featured by its flexibility: the scaling factor can be flexibly determined within a feasible scope, leading to the extension of an arbitrary context length at the inference time. Besides, Extensible Tokenization is introduced as a drop-in component, which can be seamlessly plugged into not only the LLM itself and but also its fine-tuned derivatives, bringing in the extended contextual information while fully preserving the LLM's existing capabilities. We perform comprehensive experiments on long-context language modeling and understanding tasks, which verify Extensible Tokenization as an effective, efficient, flexible, and compatible method to extend LLM's context. Our model and source code will be made publicly available.

BoolGebra: Attributed Graph-learning for Boolean Algebraic Manipulation

Boolean algebraic manipulation is at the core of logic synthesis in Electronic Design Automation (EDA) design flow. Existing methods struggle to fully exploit optimization opportunities, and often suffer from an explosive search space and limited scalability efficiency. This work presents BoolGebra, a novel attributed graph-learning approach for Boolean algebraic manipulation that aims to improve fundamental logic synthesis. BoolGebra incorporates Graph Neural Networks (GNNs) and takes initial feature embeddings from both structural and functional information as inputs. A fully connected neural network is employed as the predictor for direct optimization result predictions, significantly reducing the search space and efficiently locating the optimization space. The experiments involve training the BoolGebra model w.r.t design-specific and cross-design inferences using the trained model, where BoolGebra demonstrates generalizability for cross-design inference and its potential to scale from small, simple training datasets to large, complex inference datasets. Finally, BoolGebra is integrated with existing synthesis tool ABC to perform end-to-end logic minimization evaluation w.r.t SOTA baselines.

Low Complexity Turbo SIC-MMSE Detection for Orthogonal Time Frequency Space Modulation

Recently, orthogonal time frequency space (OTFS) modulation has garnered considerable attention due to its robustness against doubly-selective wireless channels. In this paper, we propose a low-complexity iterative successive interference cancellation based minimum mean squared error (SIC-MMSE) detection algorithm for zero-padded OTFS (ZP-OTFS) modulation. In the proposed algorithm, signals are detected based on layers processed by multiple SIC-MMSE linear filters for each sub-channel, with interference on the targeted signal layer being successively canceled either by hard or soft information. To reduce the complexity of computing individual layer filter coefficients, we also propose a novel filter coefficients recycling approach in place of generating the exact form of MMSE filter weights. Moreover, we design a joint detection and decoding algorithm for ZP-OTFS to enhance error performance. Compared to the conventional SIC-MMSE detection, our proposed algorithms outperform other linear detectors, e.g., maximal ratio combining (MRC), for ZP-OTFS with up to 3 dB gain while maintaining comparable computation complexity.

Sowing the Wind, Reaping the Whirlwind: The Impact of Editing Language Models

In the rapidly advancing field of artificial intelligence, the concept of Red-Teaming or Jailbreaking large language models (LLMs) has emerged as a crucial area of study. This approach is especially significant in terms of assessing and enhancing the safety and robustness of these models. This paper investigates the intricate consequences of such modifications through model editing, uncovering a complex relationship between enhancing model accuracy and preserving its ethical integrity. Our in-depth analysis reveals a striking paradox: while injecting accurate information is crucial for model reliability, it can paradoxically destabilize the model's foundational framework, resulting in unpredictable and potentially unsafe behaviors. Additionally, we propose a benchmark dataset NicheHazardQA to investigate this unsafe behavior both within the same and cross topical domain. This aspect of our research sheds light on how the edits, impact the model's safety metrics and guardrails. Our findings show that model editing serves as a cost-effective tool for topical red-teaming by methodically applying targeted edits and evaluating the resultant model behavior

Causal Layering via Conditional Entropy

Causal discovery aims to recover information about an unobserved causal graph from the observable data it generates. Layerings are orderings of the variables which place causes before effects. In this paper, we provide ways to recover layerings of a graph by accessing the data via a conditional entropy oracle, when distributions are discrete. Our algorithms work by repeatedly removing sources or sinks from the graph. Under appropriate assumptions and conditioning, we can separate the sources or sinks from the remainder of the nodes by comparing their conditional entropy to the unconditional entropy of their noise. Our algorithms are provably correct and run in worst-case quadratic time. The main assumptions are faithfulness and injective noise, and either known noise entropies or weakly monotonically increasing noise entropies along directed paths. In addition, we require one of either a very mild extension of faithfulness, or strictly monotonically increasing noise entropies, or expanding noise injectivity to include an additional single argument in the structural functions.

Collaborative Optimization of the Age of Information under Partial Observability

The significance of the freshness of sensor and control data at the receiver side, often referred to as Age of Information (AoI), is fundamentally constrained by contention for limited network resources. Evidently, network congestion is detrimental for AoI, where this congestion is partly self-induced by the sensor transmission process in addition to the contention from other transmitting sensors. In this work, we devise a decentralized AoI-minimizing transmission policy for a number of sensor agents sharing capacity-limited, non-FIFO duplex channels that introduce random delays in communication with a common receiver. By implementing the same policy, however with no explicit inter-agent communication, the agents minimize the expected AoI in this partially observable system. We cater to the partial observability due to random channel delays by designing a bootstrap particle filter that independently maintains a belief over the AoI of each agent. We also leverage mean-field control approximations and reinforcement learning to derive scalable and optimal solutions for minimizing the expected AoI collaboratively.

Balanced Graph Structure Information for Brain Disease Detection

Analyzing connections between brain regions of interest (ROI) is vital to detect neurological disorders such as autism or schizophrenia. Recent advancements employ graph neural networks (GNNs) to utilize graph structures in brains, improving detection performances. Current methods use correlation measures between ROI's blood-oxygen-level-dependent (BOLD) signals to generate the graph structure. Other methods use the training samples to learn the optimal graph structure through end-to-end learning. However, implementing those methods independently leads to some issues with noisy data for the correlation graphs and overfitting problems for the optimal graph. In this work, we proposed Bargrain (balanced graph structure for brains), which models two graph structures: filtered correlation matrix and optimal sample graph using graph convolution networks (GCNs). This approach aims to get advantages from both graphs and address the limitations of only relying on a single type of structure. Based on our extensive experiment, Bargrain outperforms state-of-the-art methods in classification tasks on brain disease datasets, as measured by average F1 scores.

IoT Cloud RAN Testbed for Ultra-Precise TDoA-based Localization in LPWANs

There have been many research efforts in the area of localization in recent years. Especially within the Internet of Things (IoT), the knowledge of position information for individual components is of great interest, for example, in asset tracking, to name just one. However, many of these use cases require a high energy efficiency, making a GNSS-based approach infeasible. One promising candidate can be found in Low Power Wide Area Networks (LPWAN), which enable battery lifetimes of up to 20 years. However, no gold standard for localization exists for these types of networks. Our work proposes a testbed architecture that allows the investigation and development of localization algorithms within LPWA Networks. The concept is built on a Cloud Radio Access Network (CRAN) architecture that allows the streaming of IQ from remote base stations to a central processing unit. Furthermore, the architecture is expanded by a synchronization concept based on Signals of Opportunity (SoO) to enable the testbed for runtime-based positioning. Therefore, we propose a hardware concept consisting of antennas and a low-cost off-the-shelf software-defined radio (SDR)-based frontend architecture and a software framework using a hypertext transfer protocol (HTTP)-based server and client architecture. The proposed system is installed in an urban environment. Initial measurements are conducted, where it can be shown that the proposed architecture can be used for highly precise Time Difference of Arrival (TDoA) measurements, offering the possibility of time synchronization down to approximately 200 ps and frequency synchronization of 3 mHz.