Tootfinder

RADAR: Benchmarking Language Models on Imperfect Tabular Data
Ken Gu, Zhihan Zhang, Kate Lin, Yuwei Zhang, Akshay Paruchuri, Hong Yu, Mehran Kazemi, Kumar Ayush, A. Ali Heydari, Maxwell A. Xu, Girish Narayanswamy, Yun Liu, Ming-Zher Poh, Yuzhe Yang, Mark Malhotra, Shwetak Patel, Hamid Palangi, Xuhai Xu, Daniel McDuff, Tim Althoff, Xin Liu
https://

RADAR: Benchmarking Language Models on Imperfect Tabular Data
Language models (LMs) are increasingly being deployed to perform autonomous data analyses. However, their data awareness -- the ability to recognize, reason over, and appropriately handle data artifacts such as missing values, outliers, and logical inconsistencies -- remains underexplored. These artifacts are especially common in real-world tabular data and, if mishandled, can significantly compromise the validity of analytical conclusions. To address this gap, we present RADAR, a benchmark for…

Calamus 28 When I peruse the conquered fame of heroes
My surface reading is this is a celebration of the common man, perhaps soldiers, over the more famous generals and presidents. But this commentary encourages me to dig deeper.
First, to highlight the gay text...
the brotherhood of lovers ... Through youth, and through middle and old age, how unfaltering, how affectionate and faithful they were
Very homoromantic language! Male lovers who stay together through their whole lives, affectionate.
The last fillip here is Whitman's own stance: "pensive... filled with the bitterest envy". Whitman admires these lovers and envies them. That's a striking feeling to disclose!

This https://arxiv.org/abs/2505.01903 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csLG_…

LookAlike: Consistent Distractor Generation in Math MCQs
Large language models (LLMs) are increasingly used to generate distractors for multiple-choice questions (MCQs), especially in domains like math education. However, existing approaches are limited in ensuring that the generated distractors are consistent with common student errors. We propose LookAlike, a method that improves error-distractor consistency via preference optimization. Our two main innovations are: (a) mining synthetic preference pairs from model inconsistencies, and (b) alternati…

I'm a #language fiend too. Lasswell taps into a big sore point for me as well. If you read this, stay away from the comment section. You'll go nuts!
Opinion | The phrase ‘begs the question’ is begging for oblivion - The Washington Post

This common phrase is widely misused. Is it worth salvaging?
When a phrase has been misused into uselessness, maybe avoiding it is the best policy.

Zambia's pre-colonial writing system invented by women but almost killed by colonialism - BBC News
https://www.bbc.co.uk/news/articles/c5ye50xgw8vo

Zambia's lost language invented by women but almost killed by colonialism
How social media is pushing back against the still common idea that African societies never had a knowledge system.

I feel like ChatGPT is taking a jab at other types of models here.

SafeGenBench: A Benchmark Framework for Security Vulnerability Detection in LLM-Generated Code
Xinghang Li, Jingzhe Ding, Chao Peng, Bing Zhao, Xiang Gao, Hongwan Gao, Xinchen Gu
https://arxiv.org/abs/2506.05692

SafeGenBench: A Benchmark Framework for Security Vulnerability Detection in LLM-Generated Code
The code generation capabilities of large language models(LLMs) have emerged as a critical dimension in evaluating their overall performance. However, prior research has largely overlooked the security risks inherent in the generated code. In this work, we introduce \benchmark, a benchmark specifically designed to assess the security of LLM-generated code. The dataset encompasses a wide range of common software development scenarios and vulnerability types. Building upon this benchmark, we deve…

Diarization-Aware Multi-Speaker Automatic Speech Recognition via Large Language Models
Yuke Lin, Ming Cheng, Ze Li, Beilong Tang, Ming Li
https://arxiv.org/abs/2506.05796

Diarization-Aware Multi-Speaker Automatic Speech Recognition via Large Language Models
Multi-speaker automatic speech recognition (MS-ASR) faces significant challenges in transcribing overlapped speech, a task critical for applications like meeting transcription and conversational analysis. While serialized output training (SOT)-style methods serve as common solutions, they often discard absolute timing information, limiting their utility in time-sensitive scenarios. Leveraging recent advances in large language models (LLMs) for conversational audio processing, we propose a novel…

Common Corpus: The Largest Collection of Ethical Data for LLM Pre-Training
Pierre-Carl Langlais, Carlos Rosas Hinostroza, Mattia Nee, Catherine Arnett, Pavel Chizhov, Eliot Krzystof Jones, Ir\`ene Girard, David Mach, Anastasia Stasenko, Ivan P. Yamshchikov
https://arxiv.org/abs/2506.01732

Common Corpus: The Largest Collection of Ethical Data for LLM Pre-Training
Large Language Models (LLMs) are pre-trained on large amounts of data from different sources and domains. These data most often contain trillions of tokens with large portions of copyrighted or proprietary content, which hinders the usage of such models under AI legislation. This raises the need for truly open pre-training data that is compliant with the data security regulations. In this paper, we introduce Common Corpus, the largest open dataset for language model pre-training. The data assem…

I'm a #language fiend too. Lasswell taps into a big sore point for me as well. If you read this, stay away from the comment section. You'll go nuts!
Opinion | The phrase ‘begs the question’ is begging for oblivion - The Washington Post

This common phrase is widely misused. Is it worth salvaging?
When a phrase has been misused into uselessness, maybe avoiding it is the best policy.

A Dynamic Framework for Semantic Grouping of Common Data Elements (CDE) Using Embeddings and Clustering
Madan Krishnamurthy, Daniel Korn, Melissa A Haendel, Christopher J Mungall, Anne E Thessen
https://arxiv.org/abs/2506.02160

A Dynamic Framework for Semantic Grouping of Common Data Elements (CDE) Using Embeddings and Clustering
This research aims to develop a dynamic and scalable framework to facilitate harmonization of Common Data Elements (CDEs) across heterogeneous biomedical datasets by addressing challenges such as semantic heterogeneity, structural variability, and context dependence to streamline integration, enhance interoperability, and accelerate scientific discovery. Our methodology leverages Large Language Models (LLMs) for context-aware text embeddings that convert CDEs into dense vectors capturing semant…

SafeCOMM: What about Safety Alignment in Fine-Tuned Telecom Large Language Models?
Aladin Djuhera, Swanand Ravindra Kadhe, Farhan Ahmed, Syed Zawad, Holger Boche, Walid Saad
https://arxiv.org/abs/2506.00062

SafeCOMM: What about Safety Alignment in Fine-Tuned Telecom Large Language Models?
Fine-tuning large language models (LLMs) for telecom tasks and datasets is a common practice to adapt general-purpose models to the telecom domain. However, little attention has been paid to how this process may compromise model safety. Recent research has shown that even benign fine-tuning can degrade the safety alignment of LLMs, causing them to respond to harmful or unethical user queries. In this paper, we investigate this issue for telecom-tuned LLMs using three representative datasets fea…

This https://arxiv.org/abs/2502.00722 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDC_…

Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUs
Recent advancements in Large Language Models (LLMs) have led to increasingly diverse requests, accompanied with varying resource (compute and memory) demands to serve them. However, this in turn degrades the cost-efficiency of LLM serving as common practices primarily rely on homogeneous GPU resources. In response to this problem, this work conducts a thorough study about serving LLMs over heterogeneous GPU resources on cloud platforms. The rationale is that different GPU types exhibit distinct…

To Embody or Not: The Effect Of Embodiment On User Perception Of LLM-based Conversational Agents
Kyra Wang, Boon-Kiat Quek, Jessica Goh, Dorien Herremans
https://arxiv.org/abs/2506.02514

To Embody or Not: The Effect Of Embodiment On User Perception Of LLM-based Conversational Agents
Embodiment in conversational agents (CAs) refers to the physical or visual representation of these agents, which can significantly influence user perception and interaction. Limited work has been done examining the effect of embodiment on the perception of CAs utilizing modern large language models (LLMs) in non-hierarchical cooperative tasks, a common use case of CAs as more powerful models become widely available for general use. To bridge this research gap, we conducted a mixed-methods withi…

Fast-in-Slow: A Dual-System Foundation Model Unifying Fast Manipulation within Slow Reasoning
Hao Chen, Jiaming Liu, Chenyang Gu, Zhuoyang Liu, Renrui Zhang, Xiaoqi Li, Xiao He, Yandong Guo, Chi-Wing Fu, Shanghang Zhang, Pheng-Ann Heng
https://arxiv.org/abs/2506.01953

Fast-in-Slow: A Dual-System Foundation Model Unifying Fast Manipulation within Slow Reasoning
Generalized policy and execution efficiency constitute the two critical challenges in robotic manipulation. While recent foundation policies benefit from the common-sense reasoning capabilities of internet-scale pretrained vision-language models (VLMs), they often suffer from low execution frequency. To mitigate this dilemma, dual-system approaches, inspired by Kahneman's theory, have been proposed to leverage a VLM-based System 2 model handling high-level reasoning and a separate System 1 acti…

Found in Naomi Klein's Doppelganger. The one lesson I wish the modern left would internalize more than anything.
Use shared language.
Don't use lefty jargon when the person you're talking to doesn't subscribe to these ideas. No white privilege, no colonial oppression, no intersectionality. Translate to a shared language, built on reasonable, common assumptions and do the work required to bring people on board.
Because you can be damned sure the MAGA crowd ar…

This https://arxiv.org/abs/2504.09642 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_ees…

HBS -- Hardware Build System: A Tcl-based, minimal common abstraction approach for build system for hardware designs
Build systems become an indispensable part of the software implementation and deployment process. New programming languages are released with the build system integrated into the language tools, for example, Go, Rust, or Zig. However, in the hardware description domain, no official build systems have been released with the predominant Hardware Description Languages (HDL) such as VHDL or SystemVerilog. Moreover, hardware design projects are often multilanguage. The paper proposes a new build s…

Linguist Noam Chomsky is interviewed by Common Dreams writer C.J. Polychroniou on the subject of ChatGPT in this May 2023 piece. Chomsky's stance on LLMs is that as long as we're not able to atomically understand what goes on in the statistical black box that currently is an LLM, linguists won't be able to benefit from being able to see whether it learns language like a human does, or not.
"Noam Chomsky Speaks on What ChatGPT Is Really Good For"

This https://arxiv.org/abs/2505.22251 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_ees…

Evaluation of LLMs in Speech is Often Flawed: Test Set Contamination in Large Language Models for Speech Recognition
Recent work suggests that large language models (LLMs) can improve performance of speech tasks compared to existing systems. To support their claims, results on LibriSpeech and Common Voice are often quoted. However, this work finds that a substantial amount of the LibriSpeech and Common Voice evaluation sets appear in public LLM pretraining corpora. This calls into question the reliability of findings drawn from these two datasets. To measure the impact of contamination, LLMs trained with or w…

Fairness Dynamics During Training
Krishna Patel, Nivedha Sivakumar, Barry-John Theobald, Luca Zappella, Nicholas Apostoloff
https://arxiv.org/abs/2506.01709

Fairness Dynamics During Training
We investigate fairness dynamics during Large Language Model (LLM) training to enable the diagnoses of biases and mitigations through training interventions like early stopping; we find that biases can emerge suddenly and do not always follow common performance metrics. We introduce two new metrics to evaluate fairness dynamics holistically during model pre-training: Average Rank and Jensen-Shannon Divergence by Parts. These metrics provide insights into the Pythia models' progression of biases…

This https://arxiv.org/abs/2505.11582 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csIR_…

Comparing Lexical and Semantic Vector Search Methods When Classifying Medical Documents
Classification is a common AI problem, and vector search is a typical solution. This transforms a given body of text into a numerical representation, known as an embedding, and modern improvements to vector search focus on optimising speed and predictive accuracy. This is often achieved through neural methods that aim to learn language semantics. However, our results suggest that these are not always the best solution. Our task was to classify rigidly-structured medical documents according to t…

Prompt Engineer: Analyzing Skill Requirements in the AI Job Market
An Vu, Jonas Oppenlaender
https://arxiv.org/abs/2506.00058 https://

Prompt Engineer: Analyzing Skill Requirements in the AI Job Market
The rise of large language models (LLMs) has created a new job role: the Prompt Engineer. Despite growing interest in this position, we still do not fully understand what skills this new job role requires or how common these jobs are. We analyzed 20,662 job postings on LinkedIn, including 72 prompt engineer positions, to learn more about this emerging role. We found that prompt engineering is still rare (less than 0.5% of sampled job postings) but has a unique skill profile. Prompt engineers ne…

This https://arxiv.org/abs/2407.00066 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDC_…

Compress then Serve: Serving Thousands of LoRA Adapters with Little Overhead
Fine-tuning large language models (LLMs) with low-rank adaptations (LoRAs) has become common practice, often yielding numerous copies of the same LLM differing only in their LoRA updates. This paradigm presents challenges for systems that serve real-time responses to queries that each involve a different LoRA. Prior works optimize the design of such systems but still require continuous loading and offloading of LoRAs, as it is infeasible to store thousands of LoRAs in GPU memory. To mitigate th…

Conversational Alignment with Artificial Intelligence in Context
Rachel Katharine Sterken (University of Hong Kong), James Ravi Kirkpatrick (University of Oxford,Magdalen College, Oxford)
https://arxiv.org/abs/2505.22907

Conversational Alignment with Artificial Intelligence in Context
The development of sophisticated artificial intelligence (AI) conversational agents based on large language models raises important questions about the relationship between human norms, values, and practices and AI design and performance. This article explores what it means for AI agents to be conversationally aligned to human communicative norms and practices for handling context and common ground and proposes a new framework for evaluating developers' design choices. We begin by drawing on th…