semantics

Steinert-Threlkeld and Szymanik argue that machine learning models can serve as 'cognitive thermometers' for measuring semantic complexity, showing that learning-based measures converge with and sometimes better explain semantic universals than traditional logical complexity approaches.

This paper examines the syntax and semantics of goal representations, analyzing their compositionality and role in rational behavior for cognitive science and AI.

The paper proposes SAMA-ASR, a multimodal adapter that improves automatic speech recognition for low-resource languages by using semantic anchors from translations and acoustic anchors from speech, with experiments on Taiwanese Hokkien and Hakka showing effectiveness over baselines.

An introductory guide to computational semantics in artificial intelligence systems, covering foundational concepts and applications.

This paper presents a design space exploration of straight-line asynchrony in programming languages, examining how async/await implementations vary across languages and their semantic consequences.

This paper formalizes a claim-replay layer for AI evaluation artifacts and censuses evaluation units, finding that most stop before deterministic inference due to missing historical evidence or semantic grounding.

This article explores programming language semantics and its critical role in memory safety, explaining operational, denotational, and axiomatic approaches in an educational context.

This paper proposes a second parameter, phase, in language interpretation, arguing that current transformer models lack explicit representation for phase, which is crucial for phenomena like allusion and irony, and suggests new architectures with agent-indexed semantic states.

This paper examines intensional anaphora, arguing that description-based presuppositions better account for pronoun licensing than existing value-based accounts, and formalizes the proposal in a new logic called Plural Intensional Presuppositional predicate calculus (PIP).

This paper compares human and LLM scalar judgments for sentences with focus particles 'even' and 'only' across different response scale configurations, finding stable semantic-driven differences but noting the model's lack of response variability.

An article explaining the fundamental difference between buttons and links in web design, focusing on their semantics: links represent destinations with re-contextualization options, while buttons represent actions bound to a specific context. The post also introduces the Button Actions proposal for HTML.

This paper measures how much formal semantic structure explains human label variation in natural language inference (NLI) using ChaosNLI data, finding group-level effects on entropy but item-level ceilings and null composition effects.

Proposes SERAF, a multimodal retrieval-augmented framework for time series forecasting that uses both numerical similarity and self-generated textual descriptions to retrieve historical patterns, improving forecasting under non-stationarity. Experiments on seven real-world datasets show effectiveness over state-of-the-art baselines.

This paper introduces a semantic-timescale analysis pipeline to study how generic vs. specific content is distributed over time in human and AI-generated speech, revealing that autocorrelation-window measures capture temporal organization of semantics beyond static lexical distributions.

This philosophical paper argues that AI chatbot outputs are meaningful under standard theories of language, without requiring anthropomorphic assumptions about mental states or intentions.

The article explains the confusing behavior of C array types, including their decay to pointers, exceptions like sizeof and function parameters, and compares it to function types, suggesting a mental model where arrays and pointers are strictly separated.

A blog post by Ben Meyer shares new-to-me facts about the HTML <dl> element, including that multiple <dd> can follow a <dt>, grouping within <div>, ARIA labeling, and that they've been called 'description lists' since 2008.

A detailed guide to using the HTML <dl>, <dt>, and <dd> elements to semantically mark up name-value pairs, with examples.

This paper studies how LLMs interpret English degree modifiers (e.g., 'slightly', 'drastically') in a controlled resource-allocation task. It finds that the model compresses 10 intensity words into 5 distinct numeric outputs, and the interpretation is heavily state-dependent, with lexical differentiation collapsing near system capacity.