Search Results for author:
Found 94 papers, 62 papers with code
Meat-Free Day Reduces Greenhouse Gas Emissions but Poses Challenges for Customer Retention and Adherence to Dietary Guidelines
Moreover, the increase in plant-based meals did not carry over to subsequent days, as evidenced by a 3. 5% rebound in animal-based meal consumption on days immediately following treated days.
Controlling Latent Diffusion Using Latent CLIP
We train Latent-CLIP on 2. 7B pairs of latent images and descriptive texts, and show that it matches zero-shot classification performance of similarly sized CLIP models on both the ImageNet benchmark and a LDM-generated version of it, demonstrating its effectiveness in assessing both real and generated content.
Generating Structured Outputs from Language Models: Benchmark and Studies
Constrained decoding has emerged as the dominant technology across sectors for enforcing structured outputs during generation.
Assessing Social Alignment: Do Personality-Prompted Large Language Models Behave Like Humans?
The ongoing revolution in language modeling has led to various novel applications, some of which rely on the emerging social abilities of large language models (LLMs).
Byte BPE Tokenization as an Inverse string Homomorphism
Tokenization is an important preprocessing step in the training and inference of large language models (LLMs).
Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers
In this paper, we address this question by analyzing latent representations (latents) during a word translation task in transformer-based LLMs.
Controllable Context Sensitivity and the Knob Behind It
In this paper, we search for a knob which controls this sensitivity, determining whether language models answer from the context or their prior knowledge.
Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks
Magentic-One uses a multi-agent architecture where a lead agent, the Orchestrator, plans, tracks progress, and re-plans to recover from errors.
Unpacking SDXL Turbo: Interpreting Text-to-Image Models with Sparse Autoencoders
We investigated the possibility of using SAEs to learn interpretable features for a few-step text-to-image diffusion models, such as SDXL Turbo.
Activation Scaling for Steering and Interpreting Language Models
Given the prompt "Rome is in", can we steer a language model to flip its prediction of an incorrect token "France" to a correct token "Italy" by only multiplying a few relevant activation vectors with scalars?
Entity Insertion in Multilingual Linked Corpora: The Case of Wikipedia
Links are a fundamental part of information networks, turning isolated pieces of knowledge into a network of information that is much richer than the sum of its parts.
Could ChatGPT get an Engineering Degree? Evaluating Higher Education Vulnerability to AI Assistants
We investigate the potential scale of this vulnerability by measuring the degree to which AI assistants can complete assessment questions in standard university-level STEM courses.
A Logical Fallacy-Informed Framework for Argument Generation
Despite the remarkable performance of Large Language Models (LLMs) in natural language processing tasks, they still struggle with generating logically sound arguments, resulting in potential risks such as spreading misinformation.
Self-Recognition in Language Models
Instead, our results suggest that given a set of alternatives, LMs seek to pick the "best" answer, regardless of its origin.
The AI Alignment Paradox
The field of AI alignment aims to steer AI systems toward human goals, preferences, and ethical principles.
Fleet of Agents: Coordinated Problem Solving with Large Language Models
While numerous frameworks have been developed to enhance the reasoning abilities of large language models (LLMs), there is a scarcity of methods that effectively balance the trade-off between cost and quality.
Edisum: Summarizing and Explaining Wikipedia Edits at Scale
To overcome this problem and help editors write useful edit summaries, we propose a model for recommending edit summaries generated by a language model trained to produce good edit summaries given the representation of an edit diff.
Can Language Models Recognize Convincing Arguments?
The capabilities of large language models (LLMs) have raised concerns about their potential to create and propagate convincing narratives.
Agentic AI: The Era of Semantic Decoding
We refer to these orchestrated interactions among semantic processors, optimizing and searching in semantic space, as semantic decoding algorithms.
Emojinize: Enriching Any Text with Emoji Translations
Emoji have become ubiquitous in written communication, on the Web and beyond.
Getting Serious about Humor: Crafting Humor Datasets with Unfunny Large Language Models
Humor is a fundamental facet of human cognition and interaction.
Making Reasoning Matter: Measuring and Improving Faithfulness of Chain-of-Thought Reasoning
In this paper, we perform a causal mediation analysis on twelve LLMs to examine how intermediate reasoning steps generated by the LLM influence the final outcome and find that LLMs do not reliably use their intermediate reasoning steps when generating an answer.
Symbolic Autoencoding for Self-Supervised Sequence Learning
Traditional language models, adept at next-token prediction in text sequences, often struggle with transduction tasks between distinct symbolic systems, particularly when parallel data is scarce.
Do Llamas Work in English? On the Latent Language of Multilingual Transformers
Tracking intermediate embeddings through their high-dimensional space reveals three distinct phases, whereby intermediate embeddings (1) start far away from output token embeddings; (2) already allow for decoding a semantically correct next token in the middle layers, but give higher probability to its version in English than in the input language; (3) finally move into an input-language-specific region of the embedding space.
Sketch-Guided Constrained Decoding for Boosting Blackbox Large Language Models without Logit Access
This paper introduces sketch-guided constrained decoding (SGCD), a novel approach to constrained decoding for blackbox LLMs, which operates without access to the logits of the blackbox LLM.
Evaluating Language Model Agency through Negotiations
We introduce an approach to evaluate language model (LM) agency using negotiation games.
A Glitch in the Matrix? Locating and Detecting Language Model Grounding with Fakepedia
We present a novel method to study grounding abilities using Fakepedia, a novel dataset of counterfactual texts constructed to clash with a model's internal parametric knowledge.
SoK: Memorization in General-Purpose Large Language Models
A major part of this success is due to their huge training datasets and the unprecedented number of model parameters, which allow them to memorize large amounts of information contained in the training data.
Prevalence and prevention of large language model use in crowd work
We show that the use of large language models (LLMs) is prevalent among crowd workers, and that targeted mitigation strategies can significantly reduce, but not eliminate, LLM use.
Stranger Danger! Cross-Community Interactions with Fringe Users Increase the Growth of Fringe Communities on Reddit
Overall, our findings suggest that curtailing fringe-interactions may reduce the growth of fringe communities on mainstream platforms.
In-class Data Analysis Replications: Teaching Students while Testing Science
Overall, we demonstrate that incorporating replication tasks into a large data science class can increase the reproducibility of scientific work as a by-product of data science instruction, thus benefiting both science and students.
Critical Evaluation of Artificial Intelligence as Digital Twin of Pathologist for Prostate Cancer Pathology
Adjusting the decision threshold for the secondary Gleason pattern from 5% to 10% improved the concordance level between pathologists and vPatho for tumor grading on prostatectomy specimens (kappa from 0. 44 to 0. 64).
Fly-Swat or Cannon? Cost-Effective Language Model Choice via Meta-Modeling
For a wide variety of tasks, inputs can be phrased as natural language prompts for an LM, from whose output the solution can then be extracted.
Flows: Building Blocks of Reasoning and Collaborating AI
To support rapid and rigorous research, we introduce the aiFlows library embodying Flows.
Artificial Artificial Artificial Intelligence: Crowd Workers Widely Use Large Language Models for Text Production Tasks
With the widespread adoption of LLMs, human gold--standard annotations are key to understanding the capabilities of LLMs and the validity of their results.
Generating Faithful Synthetic Data with Large Language Models: A Case Study in Computational Social Science
Large Language Models (LLMs) have democratized synthetic data generation, which in turn has the potential to simplify and broaden a wide gamut of NLP tasks.
Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning
In this work, we demonstrate that formal grammars can describe the output space for a much wider range of tasks and argue that GCD can serve as a unified framework for structured NLP tasks in general.
REFINER: Reasoning Feedback on Intermediate Representations
Language models (LMs) have recently shown remarkable performance on reasoning tasks by explicitly generating intermediate inferences, e. g., chain-of-thought prompting.
Exploiting Asymmetry for Synthetic Training Data Generation: SynthIE and the Case of Information Extraction
This work shows that useful data can be synthetically generated even for tasks that cannot be solved directly by LLMs: for problems with structured outputs, it is possible to prompt an LLM to perform the task in the reverse direction, by generating plausible input text for a target output structure.
Evolutionary Dynamics in a Varying Environment: Continuous versus Discrete Noise
Here, we study population dynamics subject to a fluctuating environment modeled by a varying carrying capacity changing continuously in time according to either binary random switches, or by being driven by a noise of continuous range.
Language Model Decoding as Likelihood-Utility Alignment
Specifically, by analyzing the correlation between the likelihood and the utility of predictions across a diverse set of tasks, we provide empirical evidence supporting the proposed taxonomy and a set of principles to structure reasoning when choosing a decoding algorithm.
An Ordinal Latent Variable Model of Conflict Intensity
It is based only on the action category ("what") and disregards the subject ("who") and object ("to whom") of an event, as well as contextual information, like associated casualty count, that should contribute to the perception of an event's "intensity".
Distribution inference risks: Identifying and mitigating sources of leakage
We identify three sources of leakage: (1) memorizing specific information about the $\mathbb{E}[Y|X]$ (expected label given the feature values) of interest to the adversary, (2) wrong inductive bias of the model, and (3) finiteness of the training data.
The Glass Ceiling of Automatic Evaluation in Natural Language Generation
Automatic evaluation metrics capable of replacing human judgments are critical to allowing fast development of new methods.
ESC-Rules: Explainable, Semantically Constrained Rule Sets
We describe a novel approach to explainable prediction of a continuous variable based on learning fuzzy weighted rules.
United States Politicians' Tone Became More Negative with 2016 Primary Campaigns
There is a widespread belief that the tone of US political language has become more negative recently, in particular when Donald Trump entered politics.
Quote Erat Demonstrandum: A Web Interface for Exploring the Quotebank Corpus
The use of attributed quotes is the most direct and least filtered pathway of information propagation in news.
Strong Heuristics for Named Entity Linking
Named entity linking (NEL) in news is a challenging endeavour due to the frequency of unseen and emerging entities, which necessitates the use of unsupervised or zero-shot methods.
Efficient Entity Candidate Generation for Low-Resource Languages
We also propose a light-weight and simple solution based on the construction of indexes whose design is motivated by more complex transfer learning based neural approaches.
Descartes: Generating Short Descriptions of Wikipedia Articles
Wikipedia is one of the richest knowledge sources on the Web today.
A Critical Re-evaluation of Neural Methods for Entity Alignment
Our findings shed light on the potential problems resulting from an impulsive application of neural methods as a panacea for all data analytics tasks.
On the Context-Free Ambiguity of Emoji
We find that with 30 annotations per emoji, 16 emojis (1. 2%) are completely unambiguous, whereas 55 emojis (4. 3%) are so ambiguous that their descriptions are indistinguishable from randomly chosen descriptions.
Homepage2Vec: Language-Agnostic Website Embedding and Classification
We make publicly available the curated Curlie dataset aligned across languages, the pre-trained Homepage2Vec model, and libraries
GenIE: Generative Information Extraction
Structured and grounded representation of text is typically formalized by closed information extraction, the problem of extracting an exhaustive set of (subject, relation, object) triplets that are consistent with a predefined set of entities and relations from a knowledge base schema.
Ranked #2 on
Relation Extraction
on REBEL
Better than Average: Paired Evaluation of NLP Systems
Evaluation in NLP is usually done by comparing the scores of competing systems independently averaged over a common set of test instances.
Invariant Language Modeling
In particular, we adapt a game-theoretic formulation of IRM (IRM-games) to language models, where the invariance emerges from a specific training schedule in which all the environments compete to optimize their own environment-specific loss by updating subsets of the model in a round-robin fashion.
Classifying Dyads for Militarized Conflict Analysis
We do this by devising a set of textual and graph-based features which represent each of the causes.
Laughing Heads: Can Transformers Detect What Makes a Sentence Funny?
We make progress in both respects by training and analyzing transformer-based humor recognition models on a recently introduced dataset consisting of minimal pairs of aligned sentences, one serious, the other humorous.
Low-Rank Subspaces for Unsupervised Entity Linking
We propose a light-weight and scalable entity linking method, Eigenthemes, that relies solely on the availability of entity names and a referent knowledge base.
Recursive input and state estimation: A general framework for learning from time series with missing data
Time series with missing data are signals encountered in important settings for machine learning.
Broccoli: Sprinkling Lightweight Vocabulary Learning into Everyday Information Diets
Thus, by seeing words in context, the user can assimilate new vocabulary without much conscious effort.
Are Anti-Feminist Communities Gateways to the Far Right? Evidence from Reddit and YouTube
Our results suggest that there is a large overlap between the user bases of the Alt-right and of the Manosphere and that members of the Manosphere have a bigger chance to engage with far right content than carefully chosen counterparts.
Computers and Society
Volunteer contributions to Wikipedia increased during COVID-19 mobility restrictions
Wikipedia, the largest encyclopedia ever created, is a global initiative driven by volunteer contributions.
Computers and Society
Formation of Social Ties Influences Food Choice: A Campus-Wide Longitudinal Study
To estimate causal effects from the passively observed log data, we control confounds in a matched quasi-experimental design: we identify focal users who at first do not have any regular eating partners but then start eating with a fixed partner regularly, and we match focal users into comparison pairs such that paired users are nearly identical with respect to covariates measured before acquiring the partner, where the two focal users' new eating partners diverge in the healthiness of their respective food choice.
YouNiverse: Large-Scale Channel and Video Metadata from English-Speaking YouTube
YouTube plays a key role in entertaining and informing people around the globe.
Time Series Analysis
Social and Information Networks
Computers and Society
Node Attribute Completion in Knowledge Graphs with Multi-Relational Propagation
The existing literature on knowledge graph completion mostly focuses on the link prediction task.
KLearn: Background Knowledge Inference from Summarization Data
The goal of text summarization is to compress documents to the relevant information while excluding background information already known to the receiver.
Crosslingual Topic Modeling with WikiPDA
We present Wikipedia-based Polyglot Dirichlet Allocation (WikiPDA), a crosslingual topic model that learns to represent Wikipedia articles written in any language as distributions over a common set of language-independent topics.
Adoption of Twitter's New Length Limit: Is 280 the New 140?
The prevalence of tweets around 140 characters before the switch in a given language is strongly correlated with the prevalence of tweets around 280 characters after the switch in the same language, and very long tweets are vastly more popular on Web clients than on mobile clients.
Experts and authorities receive disproportionate attention on Twitter during the COVID-19 crisis
Timely access to accurate information is crucial during the COVID-19 pandemic.
Social and Information Networks
Calibration of Google Trends Time Series
In the offline preprocessing phase, an "anchor bank" is constructed, a set of queries spanning the full spectrum of popularity, all calibrated against a common reference query by carefully chaining together multiple Google Trends requests.
A Ladder of Causal Distances
Causal discovery, the task of automatically constructing a causal model from data, is of major significance across the sciences.
On the Limitations of Cross-lingual Encoders as Exposed by Reference-Free Machine Translation Evaluation
We systematically investigate a range of metrics based on state-of-the-art cross-lingual semantic representations obtained with pretrained M-BERT and LASER.
Behavior Cloning in OpenAI using Case Based Reasoning
Learning from Observation (LfO), also known as Behavioral Cloning, is an approach for building software agents by recording the behavior of an expert (human or artificial) and using the recorded data to generate the required behavior.
Learning High Order Feature Interactions with Fine Control Kernels
We provide a methodology for learning sparse statistical models that use as features all possible multiplicative interactions among an underlying atomic set of features.
WikiHist.html: English Wikipedia's Full Revision History in HTML Format
Wikipedia is written in the wikitext markup language.
Computers and Society
Quantifying Engagement with Citations on Wikipedia
Wikipedia, the free online encyclopedia that anyone can edit, is one of the most visited sites on the Web and a common source of information for many users.
Computers and Society
Robust Cross-lingual Embeddings from Parallel Sentences
Recent advances in cross-lingual word embeddings have primarily relied on mapping-based methods, which project pretrained word embeddings from different languages into a shared space through a linear transformation.
Cross-Lingual Document Classification
Cross-Lingual Word Embeddings
+8
Deep Learning for Prostate Pathology
For case usage, these models were applied for the annotation tasks in clinician-oriented pathology reports for prostatectomy specimens.
RISE and DISE: Two Frameworks for Learning from Time Series with Missing Data
The most successful prior approaches for modeling such time series are based on recurrent neural networks that learn to impute unobserved values and then treat the imputed values as observed.
Auditing Radicalization Pathways on YouTube
Non-profits, as well as the media, have hypothesized the existence of a radicalization pipeline on YouTube, claiming that users systematically progress towards more extreme content on the platform.
Computers and Society Social and Information Networks
Privacy-Preserving Classification with Secret Vector Machines
Second, we implement SecVM's distributed framework for the Cliqz web browser and deploy it for predicting user gender in a large-scale online evaluation with thousands of clients, outperforming baselines by a large margin and thus showcasing that SecVM is suitable for production environments.
Secure Summation via Subset Sums: A New Primitive for Privacy-Preserving Distributed Machine Learning
For population studies or for the training of complex machine learning models, it is often required to gather data from different actors.
Eliciting New Wikipedia Users' Interests via Automatically Mined Questionnaires: For a Warm Welcome, Not a Cold Start
As such, these systems cannot make high-quality recommendations to newcomers without any previous interactions -- the so-called cold-start problem.
Crosslingual Document Embedding as Reduced-Rank Ridge Regression
Finally, although not trained for embedding sentences and words, it also achieves competitive performance on crosslingual sentence and word retrieval tasks.
Hot Streaks on Social Media
We show that user impact tends to have certain characteristics: First, impact is clustered in time, such that the most impactful tweets of a user appear close to each other.
Social and Information Networks
Expanding the Text Classification Toolbox with Cross-Lingual Embeddings
In particular, we test the hypothesis that embeddings with context are more effective, by multi-tasking the learning of multilingual word embeddings and text classification; we explore neural architectures for CLTC; and we move from bi- to multi-lingual word embeddings.
Reverse-Engineering Satire, or "Paper on Computational Humor Accepted Despite Making Serious Advances"
Starting from the observation that satirical news headlines tend to resemble serious news headlines, we build and analyze a corpus of satirical headlines paired with nearly identical but serious headlines.
Measuring Societal Biases from Text Corpora with Smoothed First-Order Co-occurrence
The common approach to measuring such biases using a corpus is by calculating the similarities between the embedding vector of a word (like nurse) and the vectors of the representative words of the concepts of interest (such as genders).
Churn Intent Detection in Multilingual Chatbot Conversations and Social Media
To this end, we crowdsource and publish a dataset of churn intent expressions in chatbot interactions in German and English.
Quootstrap: Scalable Unsupervised Extraction of Quotation-Speaker Pairs from Large News Corpora via Bootstrapping
We propose Quootstrap, a method for extracting quotations, as well as the names of the speakers who uttered them, from large news corpora.
Growing Wikipedia Across Languages via Recommendation
The system involves identifying missing articles, ranking the missing articles according to their importance, and recommending important missing articles to editors based on their interests.
Social and Information Networks Digital Libraries
Exploiting Social Network Structure for Person-to-Person Sentiment Analysis
Person-to-person evaluations are prevalent in all kinds of discourse and important for establishing reputations, building social bonds, and shaping public opinion.
Compressive Feature Learning
This paper addresses the problem of unsupervised feature learning for text data.
