The paper investigates how language models transform factual news into narrative stories, revealing that standard source‑agreement metrics misjudge the quality of these retellings. By having three model families rewrite 2,000 news articles as summaries and as stories, the authors measured changes across the five Ws and identified distortions—especially in place, causation, and accountability—that are often overlooked by fact‑checking. They found that stories tend to lose critical contextual details while amplifying accountability distortions, which can shift blame and remain undetected by conventional checks.
By Kwangsuk Park, JiWoong Yang, Vaniushkina Kseniia
The paper investigates how language models transform factual news into stories, revealing that standard source‑agreement metrics misjudge intentional changes. By rewriting 2,000 news articles into summaries and stories, the authors measured alterations across the five Ws, finding that stories often lose place and causation details while increasing accountability distortions—sometimes downplaying perpetrators. The study introduces ReTold‑2k, a corpus and scoring protocol to detect and mitigate such distortions.
The study investigates whether large language models retain traces of abandoned syntactic analyses after encountering garden‑path sentences. Across 24 models ranging from 124 M to 70 B parameters, the authors find that a residue of the initial, incorrect analysis influences later object attachment by about three‑quarters of the effect seen when the noun phrase is truly an object. They further show that this residue originates from role information in the noun phrase’s states computed under the initial analysis, and that adaptation toward the resolved structure scales with disambiguation error and primarily affects comprehension‑like measures.
arXiv:2608. 12630v1 Announce Type: cross Abstract: While large language models can generate entire novels, there is little information about the level of formal variation in their output over many generations.
By Mehdy Sedaghat Payam, Justin Quinn
The study investigates whether large language models retain traces of abandoned syntactic analyses after encountering garden‑path sentences. Across 24 models ranging from 124 M to 70 B parameters, the authors find that a residue of the initial, incorrect analysis influences later object attachment, while an adaptation toward the resolved structure scales with disambiguation error. Activation‑patching experiments show that the residue is carried by noun‑phrase states computed under the object analysis, and that overriding these states largely removes the effect.
By Xiang Ding
arXiv:2606. 29118v1 Announce Type: cross Abstract: Graph-based narrative extraction relies on a coherence function to score transitions between events, but the coherence metrics in current use are defined operationally and lack an information-theoretic foundation.
By Brian Keith-Norambuena
The paper audits a 366‑day autobiographical book generated by a large language model (LLM) against an independent verification corpus. Using a four‑level rubric, 354 of the 366 days (96.7%) failed verification, with only 12 days containing corroborated scenes and 19 days containing actively contradicted claims. Regenerating the same days with current models yielded 100% verification failure, while grounding the generation in the subject’s own corpus improved the rate to 83.3% but still left substantial residual failure.
By Heather Renze
arXiv:2608. 19369v1 Announce Type: new Abstract: Statistical watermarks for language models live in the freedom of the signifier: they choose among tokens that are nearly equivalent in meaning, and they are therefore eroded by exactly those transformations which move the form of a text while leaving its content in place.
By Daniele Corradetti
arXiv:2608.05576v2 Announce Type: replace
Abstract: When a large language model (LLM) writes Harry Potter fanfiction, it can produce fundamental elements of the Hogwarts universe, such as recognizabl...
By Zini Yang, Emily Wenger, Richard Jean So
arXiv:2608. 09093v1 Announce Type: cross Abstract: How a document's arrangement is written down, its notation, is a training variable that no dataset card records.
By E. M. Freeburg
The paper studies how transformer representations evolve across layers by examining the intrinsic dimensionality (ID) of token embeddings and their neighborhood structures. It finds that closed‑class tokens expand and collapse earlier than open‑class tokens, and that these changes are linked to shifts in local geometry. The authors compare encoder and decoder models, showing distinct layer‑wise behaviors, and demonstrate that geometric features alone can predict a token’s part‑of‑speech and reveal how semantic content changes across layers.
By Samuele Vallisa, Federico Ravenda, Claudio Palominos, Rui He, Andrea Raballo, Antonietta Mira, Philipp Homan, Wolfram Hinzen
The paper introduces Coupled Usage–Sense Processes (CUSP), a method that models lexical semantic change by coupling contextual distributions through latent usage components and using Markov composition to link adjacent time periods. CUSP quantifies change magnitude and timing, separates variation into component movement and internal reorganization, and attributes changes to specific transported component pairs. The approach is validated on synthetic data, English and German corpora, and a large corpus of US court opinions, providing detailed, text‑grounded insights into how word meanings evolve over time.
By Haruka Ezoe, Ryohei Hisano