Perplexity

Timeline

1. Oct 19, 2026 · Open-source release · 1 source
   sgl-project/sglang v0.5.22
   | Model | Type | Cookbook |
2. Oct 18, 2026 · Opinion / analysis · 1 source
   TinyFish Makes Search Free: What It Really Costs an AI Agent to Read the Web
   An AI agent that reads the web pays 2 kinds of cost.
3. Oct 17, 2026 · Opinion / analysis · 1 source
   20 Jev Alternatives: The Decision Models Taking On TypeSafe in 2026
   TypeSafe launched Jev on September 15, 2026.
4. Oct 17, 2026 · Open-source release · 1 source
   Perplexity Open-Sources SPACE, a Firecracker MicroVM Sandbox for AI Agents
   Perplexity has open-sourced SPACE, the sandbox platform behind every task in Perplexity Computer.
5. Oct 16, 2026 · Research paper · 1 source
   FuTransformer: FP8 prefill and INT4 decode for Llama-style inference on Hopper
   We present FUTRANSFORMER, a CUDA implementation of a Llama-style Transformer block for NVIDIA Hopper GPUs.
6. Oct 16, 2026 · Benchmark result · 1 source
   Continuous accountability for Verified bots
   The same is true for bots: verifying who operates it creates accountability, but trust depends on what it does next.
7. Oct 16, 2026 · Research paper · 1 source
   Watermarkability: Measuring and Improving Language-Model Compatibility with Logit-Based Watermarks
   We introduce watermarkability to characterize the compatibility between a model and the token sets requested by a watermark.
8. Oct 16, 2026 · Research paper · 1 source
   Causal Latent Workspaces for Looped Transformers
   We introduce the causal latent workspace Transformer: an anchored, four-round shared-depth architecture in which each position writes gated vectors into prefix-specific latent slots and retrieves a content-addressed summary before the next shared block application.
9. Oct 16, 2026 · Research paper · 1 source
   Prune the Leaky Neurons: Mitigating Private Data Leakage in Fine-Tuned LLMs via Joint-Embedding-Guided Pruning
   Fine-tuning large language models (LLMs) on proprietary records enables domain adaptation but introduces two challenges: (i) privacy leakage of memorized private data, such as personally identifiable information (PII), and (ii) balancing private data removal with utility preservation.
10. Oct 15, 2026 · Research paper · 1 source
    First Passage Attention
    Augmenting softmax attention with forget gates typically relies on summing decay contributions along the key--query path.
11. Oct 15, 2026 · Research paper · 1 source
    When Does Uncertainty Measure RAG Hallucination Risk?
    Uncertainty can measure hallucination risk in retrieval-augmented generation (RAG), but only when retrieved evidence constrains the answer.
12. Oct 15, 2026 · Research paper · 1 source
    Pearl-Chat: Engineering a native Luganda language model from first principles in JAX and Flax NNX
    We present Pearl-Chat, an 8,473,122-parameter ($\sim$8.5M) causal decoder-only transformer engineered for Luganda (Oluganda, ISO 639-2: lg), a low-resource Niger-Congo Bantu language spoken by over 10 million people in East Africa.
13. Oct 15, 2026 · Research paper · 1 source
    Designing Transformers for Continuous Diffusion Language Models
    LangDiT significantly reduces generation perplexity in matched LangFlow comparisons, establishing a new state of the art in continuous language generation across different model formulations.
14. Oct 15, 2026 · Research paper · 1 source
    RoleQ: Role-Aligned Scales for Converting Language Models to Ternary
    We introduce RoleQ, a quantization framework that places scales by the functional roles of Transformer units rather than by the layout of the weight matrix, as conventional formats do.
15. Oct 15, 2026 · Opinion / analysis · 1 source
    I tested Qwen3.8 27B KLD for 21 Models at 9 different KV quants that fit on 16GB VRAM (189 benchmarks).
    I calculated KLD and perplexity for Qwen3.8 27B models that fit 16GB VRAM:
16. Oct 15, 2026 · Research paper · 1 source
    Calibration Language Shapes Block-Influence Pruning in Multilingual Language Models
    We study whether English calibration transfers to Indian languages in Qwen2.5-1.5B-Instruct, Llama-3.2-1B, and Gemma-2-2B.
17. Oct 14, 2026 · Research paper · 1 source
    Beyond Expert Rankings: Contrast Calibration for Mixed-Precision MoE Quantization
    To address this gap, we propose NatScale, a contrast-calibrated allocation framework for mixed-precision MoE quantization.
18. Oct 14, 2026 · Research paper · 2 sources
    DAGFormer: Global Topology Prediction for Transformer Cross-Layer Connections
    We propose DAGFormer, a Transformer language model whose cross-layer connections are planned by a jointly trained topology predictor.
19. Oct 14, 2026 · Research paper · 1 source
    What You Sign Is Not What You Run: The Verifiability Gap in Quantized LLM Supply Chains
    Model signing authenticates the full-precision weights W an LLM is released with, yet users run a quantized artifact Q whose producing transformation is not uniquely determined under current disclosure.
20. Oct 13, 2026 · Research paper · 1 source
    Spectral-MoE: Heterogeneous Per-Layer Expert Utilization in Dense-to-MoE Conversion
    We introduce Spectral-MoE, which learns to allocate a utilization budget heterogeneously across layers.
21. Oct 13, 2026 · Research paper · 1 source
    Two For The Price Of One: Curvature-Aware Pruning Preserves Influence Functions Too
    Post-training pruning enables efficient deployment, but conventional methods do not explicitly preserve internal-gradient alignment between sparse and dense models.
22. Oct 13, 2026 · Research paper · 1 source
    Multi-Stream Residual Computation in Diffusion Language Models
    Diffusion language models reconstruct masked tokens in parallel, offering an alternative to autoregressive generation.
23. Oct 13, 2026 · Opinion / analysis · 1 source
    Update: Converting dense models into Mixture-of-Experts
    I have also released GGUF versions for both the 0.5B and the 1.5B models which can be found here bayliner1980/Qwen2.5-0.5B-MoE-A0.3B-GGUF and here bayliner1980/Qwen2.5-1.5B-MoE-A0.9B-GGUF.
24. Oct 13, 2026 · Research paper · 1 source
    Same Function, Different Trajectory: No Standard Optimizer Is Equivariant to Transformer Reparameterization
    We show that two Transformers representing the same function can follow different training trajectories.
25. Oct 13, 2026 · Research paper · 1 source
    inLoRA-MINT: Internal Representations for Auditing Training Data in Domain-adapted LLMs
    This work investigates the application of the Membership Inference Test (MINT) to large language models (LLMs) fine-tuned with LoRA, aiming to determine whether specific samples were used during training and address concerns about data privacy, leakage, and transparency.
26. Oct 13, 2026 · Research paper · 1 source
    Perturbation, Not Provenance: Why Activation Steering Does Not Make a Robust Text Watermark
    Linear directions in the activation space of large language models causally and universally influence model behavior.
27. Oct 13, 2026 · Opinion / analysis · 1 source
    I clustered two old Android phones to run a 27B LLM locally. Bypassed RAM limits via C++ RPC split and Hadamard Ternary math (No overheating)
    I saw a lot of posts here struggling to run 27B models on 8GB VRAM or complaining about phones turning into space heaters.
28. Oct 12, 2026 · Research paper · 1 source
    Distilling Text LMs into Speech-to-Speech LMs by Matching Internal Text Predictions
    Recent work has shown that knowledge distillation from pretrained text LMs offers a promising way to transfer their linguistic and factual capabilities to speech models.
29. Oct 12, 2026 · Opinion / analysis · 1 source
    llama.cpp optimizations for specific hardware and models
    I've made a few posts recently, just talking about the hardware I've got and the model I use.
30. Oct 12, 2026 · Research paper · 2 sources
    ProbabilityBench: Model Evaluation and Calibration with Exact Ground-Truth Uncertainty
    To address this, we introduce ProbabilityBench, a benchmark that provides exact, instance-level class probabilities across three simulated text-based domains: peer review, covert information transfer, and credit risk.

Often appears with