Computation and Language
PulseAugur coverage of Computation and Language — every cluster mentioning Computation and Language across labs, papers, and developer communities, ranked by signal.
- instance of Gotit.pub 90%
- instance of alphaXiv 70%
- authored ScienceCast 50%
- affiliated with alphaXiv 50%
- authored Gotit.pub 50%
- instance of CatalyzeX Code Finder for Papers 50%
- authored by Gotit.pub 50%
- instance of ScienceCast 50%
- developed CatalyzeX Code Finder for Papers 50%
- authored Influence Flower 50%
- authored by alphaXiv 50%
5 day(s) with sentiment data
-
Transformer model forecasts criminal court case outcomes
Researchers have developed a transformer-based time series model to analyze and forecast outcomes within the criminal legal system. This model treats court cases as event sequences, integrating textual data from clerk n…
-
New model captures authorial style variation for improved attribution
Researchers have developed a new authorship attribution model that accounts for an author's varied writing styles, leading to improved performance. This model was used to analyze how authorial style is expressed across …
-
New LLM training method integrates safety and reasoning earlier
Researchers have introduced a novel method for pretraining and mid-training large language models that integrates desirable behaviors like safety and factuality earlier in the process. This approach utilizes an existing…
-
New method automatically finds surprising side effects in LLM training data
Researchers have developed a method to automatically discover datasets that cause unexpected side effects in language models. These datasets, termed DSSEs, are designed to elicit specific behaviors, appear as normal pos…
-
New method extracts scholarly claims using entity anchoring
Researchers have developed a new method for extracting structured claims from full-text scholarly articles by anchoring them to specific scientific entities like tasks, datasets, and models. This approach aims to overco…
-
New measure quantifies ambiguity in legal and policy texts
Researchers have developed a new method to quantify ambiguity in prescriptive texts like laws and policies. This approach measures ambiguity by assessing the level of disagreement among stakeholders regarding how differ…
-
First token's influence on LLM reasoning length studied
A new research paper explores how the very first token in a language model's input can significantly influence the length of its subsequent reasoning. Researchers found that altering only the initial token drastically c…
-
Language models show unique jailbreak vulnerabilities based on training run
A new research paper from arXiv explores the reproducibility of language model behavior, specifically concerning jailbreaks. The study found that identical pretraining runs of a language model, even with the same data a…
-
AI language harder to understand due to information density, study finds
A new study published on arXiv suggests that AI-generated language is becoming harder for humans to understand due to its increased information density and optimization for goals beyond communication, such as math and c…
-
New inductive framework learns concept embeddings from cross-linguistic data
Researchers have developed a new inductive framework for learning concept embeddings, which are language-independent representations of meaning. This approach combines structural information from colexification networks…
-
Language models' generalization ability linked to linear fact readability
Researchers have identified a key factor in how well language models generalize knowledge after fine-tuning. They distinguish between "active facts," which aid generalization, and "inert facts," which do not. The study …
-
New framework meta-evaluates text generation metrics, finds simple methods competitive with LLM judges
Researchers have developed a new regression-based framework for meta-evaluating text generation metrics, moving beyond traditional correlation estimates. This approach aims to provide a more stable and comparable measur…
-
New study reveals serving-context nondeterminism in text classifiers
A new study published on arXiv investigates nondeterminism in text classifiers, revealing that factors like batch size, hardware, and inference engine can alter predictions even when the model and input text remain cons…
-
New research details spurious forgetting in language models
Researchers have identified a phenomenon in language models called spurious forgetting, where knowledge appears lost during finetuning but can actually be recovered. This forgetting can even reverse itself, with old fac…
-
New research identifies mechanistic signatures of faithful self-reporting in LLMs
Researchers have developed a method to distinguish between genuine introspection and confabulation in large language models. By training models with low-rank adapters on implicit decision tasks, they observed the emerge…
-
Biomedical translation models compressed using distillation and quantization
Researchers have explored combining knowledge distillation and quantization to compress neural machine translation models for the biomedical domain. This approach aims to create smaller, faster models for specialized tr…
-
New benchmark for detecting mental health stigma in online text released
Researchers have developed a new benchmark to automatically evaluate mental health stigma in online communication. This benchmark, which includes naturally occurring text from news and social media annotated with a deta…
-
New research tackles long-horizon decision-making for language agents · 4 sources tracked
Researchers are developing new methods to improve the long-horizon decision-making capabilities of language agents. One approach, RELACE, uses retrospective likelihood scoring to assess actions based on their plausibili…
-
AI models detect negation using non-verbal cues in dialogue
Researchers have developed a method to detect negation in dialogues by analyzing non-verbal cues, such as gaze and facial expressions, rather than relying solely on spoken words. The study utilized data from 27 intervie…
-
Linguistic structure enrichment fails to improve LLM coherence assessment
A new research paper explores whether adding syntactic and rhetorical information to text can improve the detection of incoherence in large language models. The study found that current model architectures were incompat…