Bahare Fatemi

Phitchaya Mangpo Phothilimthana, Sami Abu-El-Haija, Kaidi Cao et al. · stanford

Precise hardware performance models play a crucial role in code optimizations. They can assist compilers in making heuristic decisions or aid autotuners in identifying the optimal configuration for a given program. For example, the autotuner for XLA, a machine learning compiler, discovered 10-20% speedup on state-of-the-art models serving substantial production traffic at Google. Although there exist a few datasets for program performance prediction, they target small sub-programs such as basic blocks or kernels. This paper introduces TpuGraphs, a performance prediction dataset on full tensor programs, represented as computational graphs, running on Tensor Processing Units (TPUs). Each graph in the dataset represents the main computation of a machine learning workload, e.g., a training epoch or an inference step. Each data sample contains a computational graph, a compilation configuration, and the execution time of the graph when compiled with the configuration. The graphs in the dataset are collected from open-source machine learning programs, featuring popular model architectures, e.g., ResNet, EfficientNet, Mask R-CNN, and Transformer. TpuGraphs provides 25x more graphs than the largest graph property prediction dataset (with comparable graph sizes), and 770x larger graphs on average compared to existing performance prediction datasets on machine learning programs. This graph-level prediction task on large graphs introduces new challenges in learning, ranging from scalability, training efficiency, to model quality.

Kiran Vodrahalli, Santiago Ontanon, Nilesh Tripuraneni et al. · deepmind

We introduce Michelangelo: a minimal, synthetic, and unleaked long-context reasoning evaluation for large language models which is also easy to automatically score. This evaluation is derived via a novel, unifying framework for evaluations over arbitrarily long contexts which measure the model's ability to do more than retrieve a single piece of information from its context. The central idea of the Latent Structure Queries framework (LSQ) is to construct tasks which require a model to ``chisel away'' the irrelevant information in the context, revealing a latent structure in the context. To verify a model's understanding of this latent structure, we query the model for details of the structure. Using LSQ, we produce three diagnostic long-context evaluations across code and natural-language domains intended to provide a stronger signal of long-context language model capabilities. We perform evaluations on several state-of-the-art models and demonstrate both that a) the proposed evaluations are high-signal and b) that there is significant room for improvement in synthesizing long-context information.

Bahare Fatemi, Sami Abu-El-Haija, Anton Tsitsulin et al.

Graph neural networks (GNNs) demonstrate outstanding performance in a broad range of applications. While the majority of GNN applications assume that a graph structure is given, some recent methods substantially expanded the applicability of GNNs by showing that they may be effective even when no graph structure is explicitly provided. The GNN parameters and a graph structure are jointly learned. Previous studies adopt different experimentation setups, making it difficult to compare their merits. In this paper, we propose a benchmarking strategy for graph structure learning using a unified framework. Our framework, called Unified Graph Structure Learning (UGSL), reformulates existing models into a single model. We implement a wide range of existing models in our framework and conduct extensive analyses of the effectiveness of different components in the framework. Our results provide a clear and concise understanding of the different methods in this area as well as their strengths and weaknesses. The benchmark code is available at https://github.com/google-research/google-research/tree/master/ugsl.

Bahare Fatemi, Quentin Duval, Rohit Girdhar et al. · meta-ai

Recipe personalization through ingredient substitution has the potential to help people meet their dietary needs and preferences, avoid potential allergens, and ease culinary exploration in everyone's kitchen. To address ingredient substitution, we build a benchmark, composed of a dataset of substitution pairs with standardized splits, evaluation metrics, and baselines. We further introduce Graph-based Ingredient Substitution Module (GISMo), a novel model that leverages the context of a recipe as well as generic ingredient relational information encoded within a graph to rank plausible substitutions. We show through comprehensive experimental validation that GISMo surpasses the best performing baseline by a large margin in terms of mean reciprocal rank. Finally, we highlight the benefits of GISMo by integrating it in an improved image-to-recipe generation pipeline, enabling recipe personalization through user intervention. Quantitative and qualitative results show the efficacy of our proposed system, paving the road towards truly personalized cooking and tasting experiences.

Jonathan Pilault, Michael Galkin, Bahare Fatemi et al. · deepmind, mila

Recent work on Graph Neural Networks has demonstrated that self-supervised pretraining can further enhance performance on downstream graph, link, and node classification tasks. However, the efficacy of pretraining tasks has not been fully investigated for downstream large knowledge graph completion tasks. Using a contextualized knowledge graph embedding approach, we investigate five different pretraining signals, constructed using several graph algorithms and no external data, as well as their combination. We leverage the versatility of our Transformer-based model to explore graph structure generation pretraining tasks (i.e. path and k-hop neighborhood generation), typically inapplicable to most graph embedding methods. We further propose a new path-finding algorithm guided by information gain and find that it is the best-performing pretraining task across three downstream knowledge graph completion datasets. While using our new path-finding algorithm as a pretraining signal provides 2-3% MRR improvements, we show that pretraining on all signals together gives the best knowledge graph completion results. In a multitask setting that combines all pretraining tasks, our method surpasses the latest and strong performing knowledge graph embedding methods on all metrics for FB15K-237, on MRR and Hit@1 for WN18RRand on MRR and hit@10 for JF17K (a knowledge hypergraph dataset).

Bahare Fatemi, Jonathan Halcrow, Bryan Perozzi

Graphs are a powerful tool for representing and analyzing complex relationships in real-world applications such as social networks, recommender systems, and computational finance. Reasoning on graphs is essential for drawing inferences about the relationships between entities in a complex system, and to identify hidden patterns and trends. Despite the remarkable progress in automated reasoning with natural text, reasoning on graphs with large language models (LLMs) remains an understudied problem. In this work, we perform the first comprehensive study of encoding graph-structured data as text for consumption by LLMs. We show that LLM performance on graph reasoning tasks varies on three fundamental levels: (1) the graph encoding method, (2) the nature of the graph task itself, and (3) interestingly, the very structure of the graph considered. These novel results provide valuable insight on strategies for encoding graphs as text. Using these insights we illustrate how the correct choice of encoders can boost performance on graph reasoning tasks inside LLMs by 4.8% to 61.8%, depending on the task.

Mehran Kazemi, Bahare Fatemi, Hritik Bansal et al. · deepmind

Large language models (LLMs) are increasingly deployed in everyday applications, demanding robust general reasoning capabilities and diverse reasoning skillset. However, current LLM reasoning benchmarks predominantly focus on mathematical and coding abilities, leaving a gap in evaluating broader reasoning proficiencies. One particular exception is the BIG-Bench dataset, which has served as a crucial benchmark for evaluating the general reasoning capabilities of LLMs, thanks to its diverse set of challenging tasks that allowed for a comprehensive assessment of general reasoning across various skills within a unified framework. However, recent advances in LLMs have led to saturation on BIG-Bench, and its harder version BIG-Bench Hard (BBH). State-of-the-art models achieve near-perfect scores on many tasks in BBH, thus diminishing its utility. To address this limitation, we introduce BIG-Bench Extra Hard (BBEH), a new benchmark designed to push the boundaries of LLM reasoning evaluation. BBEH replaces each task in BBH with a novel task that probes a similar reasoning capability but exhibits significantly increased difficulty. We evaluate various models on BBEH and observe a (harmonic) average accuracy of 9.8\% for the best general-purpose model and 44.8\% for the best reasoning-specialized model, indicating substantial room for improvement and highlighting the ongoing challenge of achieving robust general reasoning in LLMs. We release BBEH publicly at: https://github.com/google-deepmind/bbeh.

Ivaxi Sheth, Bahare Fatemi, Mario Fritz

Causality is essential in scientific research, enabling researchers to interpret true relationships between variables. These causal relationships are often represented by causal graphs, which are directed acyclic graphs. With the recent advancements in Large Language Models (LLMs), there is an increasing interest in exploring their capabilities in causal reasoning and their potential use to hypothesize causal graphs. These tasks necessitate the LLMs to encode the causal graph effectively for subsequent downstream tasks. In this paper, we introduce CausalGraph2LLM, a comprehensive benchmark comprising over 700k queries across diverse causal graph settings to evaluate the causal reasoning capabilities of LLMs. We categorize the causal queries into two types: graph-level and node-level queries. We benchmark both open-sourced and propriety models for our study. Our findings reveal that while LLMs show promise in this domain, they are highly sensitive to the encoding used. Even capable models like GPT-4 and Gemini-1.5 exhibit sensitivity to encoding, with deviations of about $60\%$. We further demonstrate this sensitivity for downstream causal intervention tasks. Moreover, we observe that LLMs can often display biases when presented with contextual information about a causal graph, potentially stemming from their parametric memory.

Zhikai Chen, Haitao Mao, Jingzhe Liu et al.

Given the ubiquity of graph data and its applications in diverse domains, building a Graph Foundation Model (GFM) that can work well across different graphs and tasks with a unified backbone has recently garnered significant interests. A major obstacle to achieving this goal stems from the fact that graphs from different domains often exhibit diverse node features. Inspired by multi-modal models that align different modalities with natural language, the text has recently been adopted to provide a unified feature space for diverse graphs. Despite the great potential of these text-space GFMs, current research in this field is hampered by two problems. First, the absence of a comprehensive benchmark with unified problem settings hinders a clear understanding of the comparative effectiveness and practical value of different text-space GFMs. Second, there is a lack of sufficient datasets to thoroughly explore the methods' full potential and verify their effectiveness across diverse settings. To address these issues, we conduct a comprehensive benchmark providing novel text-space datasets and comprehensive evaluation under unified problem settings. Empirical results provide new insights and inspire future research directions. Our code and data are publicly available from \url{https://github.com/CurryTang/TSGFM}.

Mehran Kazemi, Nishanth Dikkala, Ankit Anand et al.

With the continuous advancement of large language models (LLMs), it is essential to create new benchmarks to effectively evaluate their expanding capabilities and identify areas for improvement. This work focuses on multi-image reasoning, an emerging capability in state-of-the-art LLMs. We introduce ReMI, a dataset designed to assess LLMs' ability to Reason with Multiple Images. This dataset encompasses a diverse range of tasks, spanning various reasoning domains such as math, physics, logic, code, table/chart understanding, and spatial and temporal reasoning. It also covers a broad spectrum of characteristics found in multi-image reasoning scenarios. We have benchmarked several cutting-edge LLMs using ReMI and found a substantial gap between their performance and human-level proficiency. This highlights the challenges in multi-image reasoning and the need for further research. Our analysis also reveals the strengths and weaknesses of different models, shedding light on the types of reasoning that are currently attainable and areas where future models require improvement. To foster further research in this area, we are releasing ReMI publicly: https://huggingface.co/datasets/mehrankazemi/ReMI.

Bahare Fatemi, Mehran Kazemi, Anton Tsitsulin et al.

Large language models (LLMs) have showcased remarkable reasoning capabilities, yet they remain susceptible to errors, particularly in temporal reasoning tasks involving complex temporal logic. Existing research has explored LLM performance on temporal reasoning using diverse datasets and benchmarks. However, these studies often rely on real-world data that LLMs may have encountered during pre-training or employ anonymization techniques that can inadvertently introduce factual inconsistencies. In this work, we address these limitations by introducing novel synthetic datasets specifically designed to assess LLM temporal reasoning abilities in various scenarios. The diversity of question types across these datasets enables systematic investigation into the impact of the problem structure, size, question type, fact order, and other factors on LLM performance. Our findings provide valuable insights into the strengths and weaknesses of current LLMs in temporal reasoning tasks. To foster further research in this area, we are open-sourcing the datasets and evaluation framework used in our experiments: https://huggingface.co/datasets/baharef/ToT.

Bahare Fatemi, Seyed Mehran Kazemi, David Poole

Relational logistic regression (RLR) is a representation of conditional probability in terms of weighted formulae for modelling multi-relational data. In this paper, we develop a learning algorithm for RLR models. Learning an RLR model from data consists of two steps: 1- learning the set of formulae to be used in the model (a.k.a. structure learning) and learning the weight of each formula (a.k.a. parameter learning). For structure learning, we deploy Schmidt and Murphy's hierarchical assumption: first we learn a model with simple formulae, then more complex formulae are added iteratively only if all their sub-formulae have proven effective in previous learned models. For parameter learning, we convert the problem into a non-relational learning problem and use an off-the-shelf logistic regression learning algorithm from Weka, an open-source machine learning tool, to learn the weights. We also indicate how hidden features about the individuals can be incorporated into RLR to boost the learning performance. We compare our learning algorithm to other structure and parameter learning algorithms in the literature, and compare the performance of RLR models to standard logistic regression and RDN-Boost on a modified version of the MovieLens data-set.

Bryan Perozzi, Bahare Fatemi, Dustin Zelle et al.

How can we best encode structured data into sequential form for use in large language models (LLMs)? In this work, we introduce a parameter-efficient method to explicitly represent structured data for LLMs. Our method, GraphToken, learns an encoding function to extend prompts with explicit structured information. Unlike other work which focuses on limited domains (e.g. knowledge graph representation), our work is the first effort focused on the general encoding of structured data to be used for various reasoning tasks. We show that explicitly representing the graph structure allows significant improvements to graph reasoning tasks. Specifically, we see across the board improvements - up to 73% points - on node, edge and, graph-level tasks from the GraphQA benchmark.

Kartikeya Badola, Jonathan Simon, Arian Hosseini et al.

Large language models (LLMs) excel at solving problems with clear and complete statements, but often struggle with nuanced environments or interactive tasks which are common in most real-world scenarios. This highlights the critical need for developing LLMs that can effectively engage in logically consistent multi-turn dialogue, seek information and reason with incomplete data. To this end, we introduce a novel benchmark comprising a suite of multi-turn tasks each designed to test specific reasoning, interactive dialogue, and information-seeking abilities. These tasks have deterministic scoring mechanisms, thus eliminating the need for human intervention. Evaluating frontier models on our benchmark reveals significant headroom. Our analysis shows that most errors emerge from poor instruction following, reasoning failures, and poor planning. This benchmark provides valuable insights into the strengths and weaknesses of current LLMs in handling complex, interactive scenarios and offers a robust platform for future research aimed at improving these critical capabilities.

Bahare Fatemi, Perouz Taslakian, David Vazquez et al.

Embedding-based methods for reasoning in knowledge hypergraphs learn a representation for each entity and relation. Current methods do not capture the procedural rules underlying the relations in the graph. We propose a simple embedding-based model called ReAlE that performs link prediction in knowledge hypergraphs (generalized knowledge graphs) and can represent high-level abstractions in terms of relational algebra operations. We show theoretically that ReAlE is fully expressive and provide proofs and empirical evidence that it can represent a large subset of the primitive relational algebra operations, namely renaming, projection, set union, selection, and set difference. We also verify experimentally that ReAlE outperforms state-of-the-art models in knowledge hypergraph completion, and in representing each of these primitive relational algebra operations. For the latter experiment, we generate a synthetic knowledge hypergraph, for which we design an algorithm based on the Erdos-R'enyi model for generating random graphs.

Bahare Fatemi, Layla El Asri, Seyed Mehran Kazemi

Graph neural networks (GNNs) work well when the graph structure is provided. However, this structure may not always be available in real-world applications. One solution to this problem is to infer a task-specific latent structure and then apply a GNN to the inferred graph. Unfortunately, the space of possible graph structures grows super-exponentially with the number of nodes and so the task-specific supervision may be insufficient for learning both the structure and the GNN parameters. In this work, we propose the Simultaneous Learning of Adjacency and GNN Parameters with Self-supervision, or SLAPS, a method that provides more supervision for inferring a graph structure through self-supervision. A comprehensive experimental study demonstrates that SLAPS scales to large graphs with hundreds of thousands of nodes and outperforms several models that have been proposed to learn a task-specific graph structure on established benchmarks.

Bahare Fatemi, Perouz Taslakian, David Vazquez et al.

Knowledge graphs store facts using relations between two entities. In this work, we address the question of link prediction in knowledge hypergraphs where relations are defined on any number of entities. While techniques exist (such as reification) that convert non-binary relations into binary ones, we show that current embedding-based methods for knowledge graph completion do not work well out of the box for knowledge graphs obtained through these techniques. To overcome this, we introduce HSimplE and HypE, two embedding-based methods that work directly with knowledge hypergraphs. In both models, the prediction is a function of the relation embedding, the entity embeddings and their corresponding positions in the relation. We also develop public datasets, benchmarks and baselines for hypergraph prediction and show experimentally that the proposed models are more effective than the baselines.

Bahare Fatemi, Siamak Ravanbakhsh, David Poole

Knowledge graphs are used to represent relational information in terms of triples. To enable learning about domains, embedding models, such as tensor factorization models, can be used to make predictions of new triples. Often there is background taxonomic information (in terms of subclasses and subproperties) that should also be taken into account. We show that existing fully expressive (a.k.a. universal) models cannot provably respect subclass and subproperty information. We show that minimal modifications to an existing knowledge graph completion method enables injection of taxonomic information. Moreover, we prove that our model is fully expressive, assuming a lower-bound on the size of the embeddings. Experimental results on public knowledge graphs show that despite its simplicity our approach is surprisingly effective.

Nandini Ramanan, Gautam Kunapuli, Tushar Khot et al.

We consider the problem of learning Relational Logistic Regression (RLR). Unlike standard logistic regression, the features of RLRs are first-order formulae with associated weight vectors instead of scalar weights. We turn the problem of learning RLR to learning these vector-weighted formulae and develop a learning algorithm based on the recently successful functional-gradient boosting methods for probabilistic logic models. We derive the functional gradients and show how weights can be learned simultaneously in an efficient manner. Our empirical evaluation on standard and novel data sets demonstrates the superiority of our approach over other methods for learning RLR.

Bahare Fatemi, Seyed Mehran Kazemi, David Poole

Consider the following problem: given a database of records indexed by names (e.g., name of companies, restaurants, businesses, or universities) and a new name, determine whether the new name is in the database, and if so, which record it refers to. This problem is an instance of record linkage problem and is a challenging problem because people do not consistently use the official name, but use abbreviations, synonyms, different order of terms, different spelling of terms, short form of terms, and the name can contain typos or spacing issues. We provide a probabilistic model using relational logistic regression to find the probability of each record in the database being the desired record for a given query and find the best record(s) with respect to the probabilities. Building on term-matching and translational approaches for search, our model addresses many of the aforementioned challenges and provides good results when existing baselines fail. Using the probabilities outputted by the model, we can automate the search process for a portion of queries whose desired documents get a probability higher than a trust threshold. We evaluate our model on a large real-world dataset from a telecommunications company and compare it to several state-of-the-art baselines. The obtained results show that our model is a promising probabilistic model for record linkage for names. We also test if the knowledge learned by our model on one domain can be effectively transferred to a new domain. For this purpose, we test our model on an unseen test set from the business names of the secondString dataset. Promising results show that our model can be effectively applied to unseen datasets. Finally, we study the sensitivity of our model to the statistics of datasets.

Seyed Mehran Kazemi, Bahare Fatemi, Alexandra Kim et al.

Relational probabilistic models have the challenge of aggregation, where one variable depends on a population of other variables. Consider the problem of predicting gender from movie ratings; this is challenging because the number of movies per user and users per movie can vary greatly. Surprisingly, aggregation is not well understood. In this paper, we show that existing relational models (implicitly or explicitly) either use simple numerical aggregators that lose great amounts of information, or correspond to naive Bayes, logistic regression, or noisy-OR that suffer from overconfidence. We propose new simple aggregators and simple modifications of existing models that empirically outperform the existing ones. The intuition we provide on different (existing or new) models and their shortcomings plus our empirical findings promise to form the foundation for future representations.