THE INTERNET IS THINKING
Earlier sources: Language models
Language models learn patterns in text and other representations to predict or generate sequences. Follow model releases, training methods, reasoning, evaluation, and deployment research; a benchmark result is specific to its test conditions.
SYMTRI organizes observations from public sources into a living map. Connections reflect shared subject matter; they do not establish causation or endorse a source.
Open in the map ↗462 classified signals in the fourteen-day window. These counts include signals awaiting readable evidence.
Earlier sources
Up to ten sources with readable evidence, ordered by their feed timestamp. These archive sources can be older than the live fourteen-day window. This list is a sample, not a measure of total activity. Dates distinguish paper publication, repository creation, model/dataset activity, Hacker News posts, and KEV catalog additions. “First archived” records when SYMTRI first stored the signal.
- OpenAlex · Published · First archived
Empirical Analysis of Goal Hijacking in Large Vision-Language Models via Visual Prompt Injection ↗
Abstract In this study, we explore visual prompt injection (VPI), a method that exploits the ability of large vision-language models (LVLMs) to follow instructions embedded within input images. LVLMs, such as GPT-4V, are designed to interpret and respond to visual prompts. Although this capability enables valuable applications, it introduces substantial security risks. To address this issue, we propose a new VPI method called “goal hijacking via visual prompt injection” (GHVPI). This method redirects LVLMs from their original execution task to an alternative task specified by an attacker. Our…
- OpenAlex · Published · First archived
Automating clinical information retrieval from Finnish electronic health records using large language models ↗
Abstract Clinicians often need to search for patient-specific information within electronic health records (EHRs) containing years of heterogeneous longitudinal documentation, a task that can be time-consuming, cognitively demanding, and prone to oversight. We evaluated 14 locally deployable open-source large language models for structured patient-specific clinical information retrieval using a dataset comprising 1664 expert-annotated question–answer pairs from EHR-derived records of 183 Finnish patients undergoing breast cancer screening, diagnosis, or treatment under fully offline conditions…
- Europe PMC · Published · First archived
Genome-wide-scale prediction of compound-protein interactions using foundation and language models based on three-dimensional structures of compounds and proteins. ↗
Motivation The identification of compound-protein interactions (CPIs) is crucial in the early stages of drug discovery. However, machine-learning (ML)-based methods based on one- and two-dimensional representations cannot capture important geometric information on the binding sites of CPIs, which limits their predictive accuracy. Results We propose DESTIN (dual structure-guided estimation of substance-target interaction networks), an ML framework for CPI prediction on a genome-wide scale from the three-dimensional (3D) structures of compounds and proteins. We constructed compound feature vecto
- arXiv · Published · First archived
Faithful Activation Verbalization: Reducing Hallucinations in LLM Representation Interpretation ↗
Activation verbalization methods such as Activation Oracle and Natural Language Autoencoders decode hidden representations of large language models into human-readable natural language. However, existing methods can produce incomplete or hallucinated descriptions, making their activation verbalizations difficult to trust and use reliably in practice. To this end, we introduce AVPO, a two-stage framework that first reconstructs source text from a hidden activation and then evaluates the resulting text with a separate frozen question-answering model, yielding an explicit and inspectable intermed…
- arXiv · Published · First archived
Position Aware Layer Queries for Test Time Training in Vision Language Models ↗
Test-Time Training (TTT) adapts models to incoming test samples (e.g. out-of-distribution, (OOD)) when conventional fine-tuning is infeasible. Existing TTT methods for Vision-Language Models (VLMs) create supervision from several augmented views, each requiring forward (and often backward) passes through the entire VLM, incurring substantial computational cost. We observe that one forward pass with all the intermediate layer outputs already yields far more signal than the final embedding from all augmentations. We introduce Layer Query Network (LQN), a lightweight approach that can adapt a fro…
- arXiv · Published · First archived
Fisher-Informed Recalibration for Feedback-Based On-Policy Self-Distillation of LLMs ↗
Feedback-based on-policy self-distillation has emerged as a promising approach for enabling foundation models, more specifically Large Language Models (LLMs), to learn from their own outputs under external feedback, with a single model serving as both teacher and student. However, such methods can exhibit unstable optimization, conducive to performance collapse during training. To address this limitation, we propose FIRE (Fisher-Informed REcalibration), a dual-branch framework that recalibrates the supervision applied to correct and incorrect on-policy outputs during fine-tuning. For correct r…
- arXiv · Published · First archived
EHRAdapt: Adapting Pretrained Language Models to Electronic Health Records with Semantic Priors for Rare Clinical Events ↗
Electronic health records (EHRs) encode clinical histories as (time, modality, code) tuples, whereas pretrained language models expect text tokens. Serializing them as text inflates sequence length and redundantly encodes structure. We introduce EHRAdapt, an adapter that maps tuples directly into a frozen language model's embedding space. Modality receives a learned embedding, time gaps enter through learned attention biases, and event codes receive dedicated vectors. Learning event vectors is the central challenge: clinical vocabularies are long-tailed, leaving rare events too few observation…
- Hacker News · Posted on Hacker News · First archived
Kimi K3 vs. Claude Opus 5.5: How Two Flagship LLMs Built Pokémon Emerald Worms ↗
Hacker News discussion.
- GitHub · Repository created · First archived
kaanncavdar/ratica ↗
Free, open-source offline AI PDF translator. Translate PDF books on your own computer and keep the original page layout. PDF + EPUB output. Windows, macOS, Linux.
- Hacker News · Posted on Hacker News · First archived
The Tragic Alley of the LLM ↗
Hacker News discussion.
Live window recorded . This is the ingestion observation time, not the publication or discovery date of each source.
How to assess evidence about Language models
Check each original source for its methods, release status, limitations, and supporting data. A repository announcement, a preprint, and a published study provide different kinds of evidence. SYMTRI’s subject classification helps discovery; it does not independently validate the source’s claims.
See the collection and counting methodology ↗ · Read the weekly synthesis ↗
Related threads
Explore related regions
These navigation suggestions are part of the map’s editorial structure, rather than measured evidence of a connection.