Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.
arXiv:2608. 10672v1 Announce Type: cross Abstract: Social interaction has become one of the most common uses of LLMs, yet research on emotional bonds with AI has focused largely on how users experience these systems, leaving the systems' role in relationship formation poorly understood.
arXiv:2608. 10621v1 Announce Type: new Abstract: Recent research on Large Language Model (LLM) safety has widely adopted guardrails to identify unsafe LLM outputs.
arXiv:2608. 09934v1 Announce Type: cross Abstract: Large language model (LLM) agents improve task performance by decomposing problems into role-specialized behaviors.
Diffusion-based generative video compression has emerged as a promising paradigm to improve perceptual quality, where latent frames are required to be encoded efficiently while serving as denoising conditions. However, existing methods neither carefully design reference and quality structures during latent coding nor account for the impact of frame-level quality variation on denoising procedure, which limits coding efficiency and aggravates artifact propagation during generative reconstruction.
Mobile GUI agents remain brittle when deployed to applications absent from source training. We study novel-app generalization under a limited target interaction budget and without target demonstrations.
Recent advances in model design and self-supervised audio representations have improved speech and audio understanding, yet infant-centered naturalistic recordings remain challenging due to limited labeled data, low signal-to-noise ratio, and cross-family domain shifts. We present a family-conditioned, multi-tier audio tagger that combines a LoRA-finetuned Whisper encoder with a lightweight, target-speaker-aware Transformer for long-context inference and framewise prediction across tiers.
Achieving effective self-correction, where models verify and correct their own mistakes, remains a fundamental challenge for large language models (LLMs). In this work, we propose Self-Fix Step-DPO (SFS-DPO), a reinforcement learning based, two-stage framework for step-level self-verification and self-correction.
Uncertainty quantification (UQ) methods for language models are typically evaluated on single-turn outputs, where uncertainty is attached to one generated answer. For LLM agents, however, the unit of observation is an interactive trajectory, where the model can ask clarifying questions, call tools, update state, and make intermediate decisions whose errors propagate to the final outcome.
Group alignment adapts a language model to a demographic group to produce responses that reflect the group's opinions, values, and preferences. Sycophancy, a well-documented by-product of alignment, causes the model to over-agree with the user regardless of factual and objective information.
See how RingCentral uses ChatGPT Work and Codex to accelerate AI product development and centralize operational intelligence across engineering and operations.
There are no lossless transformations of natural-language text Sophie Alpert shares her "internal policy on acceptable use of AI writing by engineers". It's a short read (supporting its own recommendations) and really good.
Stealing Reasoning Traces from Proprietary LLM APIs A vanity domain name ( stolen-thoughts. com ) for a neat paper : Anthropic, OpenAI, and Google return encrypted chain-of-thought blocks to clients that can be replayed across sessions, users, and models.
Estimating human gaze targets from images in-the-wild is an important and formidable task. Existing approaches primarily employ brittle, multi-stage pipelines that require explicit inputs, like head bounding boxes and human pose, in order to identify the subject of gaze analysis.
Specialist training beats generalist scale when forecasting financial statements. To our knowledge, no prior work jointly forecasts complete financial statements beyond one year, yet in a discounted-cash-flow valuation most firm value sits past that window.
Synthetic dialogue generation offers a way to study conversational dynamics in sensitive domains where real data are difficult to access, release, or annotate. The underlying abuse may occur online or offline: threats and coercion can appear directly in messages, while behaviours such as surveillance, isolation, stalking, and physical violence may be planned, disclosed, or referred to conversationally.
Modern reinforcement learning (RL) post-training pipelines for large language models (LLMs) increasingly combine rollout workloads across multiple domains and feedback paradigms. Prefix-aware routing improves inference efficiency through cache reuse and load balancing, but it does not control how heterogeneous rollout sessions compete for KV-cache capacity.
Safety alignment in large language models (LLMs) is largely developed in English, assuming these safeguards generalize across multilingual settings. However, this assumption remains underexplored and exposes a vulnerability in low-resource languages.
Agentic coding READMEs like CLAUDE. md grow without bound in real repositories, stopping only when the repository retires or someone rewrites the file wholesale.
Frontier LLMs are updated frequently and typically outperform their predecessors in aggregate. But aggregate gains say little about individual samples: an update can still cause sample-level regression, where a response correct under the old model becomes incorrect under the new one.
Emergent misalignment (EM) is the phenomenon where fine-tuning a language model on a narrow task leads to harmful behavior in unrelated domains. A leading mechanistic account attributes EM to persona features: latent directions acquired during pre-training that misaligned fine-tuning amplifies.