WildChat
PulseAugur coverage of WildChat — every cluster mentioning WildChat across labs, papers, and developer communities, ranked by signal.
4 day(s) with sentiment data
-
New platform MultiTurnRoute optimizes LLM routing for multi-turn conversations
Researchers have developed MultiTurnRoute, a platform designed for building and evaluating multi-turn LLM routers. This platform incorporates a routing algorithm called Chain-of-Route (CoR), which leverages dialogue his…
-
New study reveals LLM conversations lack inspectable evidence
A new arXiv paper examines user interactions with large language models (LLMs) in real-world conversations, analyzing over 489,000 response pairs from ShareChat and WildChat. The study found that while attribution langu…
-
New method detects psychological harm risk factors in LLM conversations
Researchers have developed a new method to detect potential psychological harm in conversations with large language models by focusing on observable AI behaviors rather than direct harm assessment. This approach, termed…
-
New AC3 method combats AI conversation context pollution
Researchers have developed Agentic Conversation Context Curation (AC3), a novel method to combat context pollution in multi-turn AI conversations. This training-free approach uses a separate analyzer to consolidate the …
-
AI jailbreak detection struggles with real-world multi-turn conversations
A new research paper explores the effectiveness of gradient-based methods for detecting jailbreak prompts in multi-turn AI conversations. The study found that while methods like GradSafe perform well on synthetic data, …
-
New research probes LLM robustness, explanations, and interaction methods
Researchers are exploring new methods to evaluate and understand Large Language Models (LLMs). One study introduces the SAST-IR framework to test LLMs' factual robustness against persuasion attacks, revealing a high suc…
-
New research suggests prompt phrasing significantly impacts LLM advice responses
A new research paper from the Large Model Systems Organization, published on arXiv, proposes a novel approach to understanding how the way users phrase advice-seeking prompts influences Large Language Model (LLM) respon…
-
AI Observatory provides independent analysis of AI usage, challenging company claims
A new research project called the AI Observatory aims to provide independent insights into how people are using generative AI models. Unlike reports from companies like Anthropic and OpenAI, which researchers claim sele…
-
New benchmark evaluates multi-agent routing accuracy-cost trade-offs
Researchers have introduced a new benchmark and evaluation protocol for multi-agent routing, framing it as a set-valued prediction problem. The benchmark, derived from WildChat, comprises 3,000 prompts and a catalog of …
-
Public chat data explored for AI model safety evaluation
Researchers are exploring the use of public chat data as an alternative to private production data for evaluating frontier AI models. This approach, termed Deployment Simulation, aims to predict undesirable model behavi…
-
OpenAI unveils deployment simulation to predict AI model behavior
OpenAI has developed a new method called Deployment Simulation to predict how AI models will behave in real-world scenarios before they are released. This technique uses de-identified user data to simulate deployment co…
-
Study: Commercial LLMs Outperform Open-Weight Models on Security Prompts
A new study analyzed 14,727 security and privacy prompts from the WildChat dataset, revealing that users frequently seek advice on protecting themselves online. Commercial large language models, such as GPT 5.5, demonst…
-
AI chatbots repeat Elias Thorne stories due to alignment training
A recurring character named Elias Thorne, often depicted as a lighthouse keeper or clockmaker, is appearing in a significant percentage of stories generated by various large language models. Researchers from Cornell Uni…
-
New research tackles AI agent training with realistic user personas
Two new research papers explore the limitations of current user simulators for training AI agents. The first paper introduces Persona Policies (PPol), a method to generate more realistic and varied user personas for sim…
-
New benchmark 'Prosa' evaluates LLMs on Brazilian Portuguese chats
Researchers have introduced Prosa, a new benchmark designed to evaluate Large Language Models (LLMs) using real user conversations in Brazilian Portuguese. This benchmark utilizes a rubric-based scoring system with mult…