Skip to content
Papers.

Salesforce research papers

Company · 33 papers · 0 citations · latest Sep 2026 3 days ago

Research page

33 of 33 papers, newest first

PaperCitations
Just-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM AgentsKeeps an agent's raw past runs and writes a task-specific memory only when a new task arrives, rather than deciding up front what to keep.Agents and evaluation · Sep 20263 days ago-
Flattening Every Memory Peak in Long-Context Mixture-of-Experts TrainingBounds the four memory peaks that crash long-context mixture-of-experts training, from expert dispatch to optimizer state, with fixed-size GPU schedules.Architectures · Sep 202613 days ago-
RISE: Recursive Improvement via Self-Extrapolating Policy DistillationRISE improves language model post-training by recursively generating dense token-level supervision from the model's own reinforcement learning trajectory via self-extrapolation, avoiding external teachers.Training and scaling · Sep 20263 weeks ago-
EVOHARNESSBENCH: Can Your Agents Keep Pace with an Evolving Harness?The study introduces a benchmark to evaluate LLM agents under evolving tool, skill, and agent harnesses, revealing persistent gaps in retention, adaptation, and harness-induced forgetting.Agents and evaluation · Sep 20263 weeks ago-
Random Attention: Rethinking KV Cache Eviction for Efficient ReasoningRandom eviction of reasoning tokens matches selective KV cache compression because reasoning traces are self-protecting through redundancy, making scoring unnecessary once prompts are preserved.Reasoning · Sep 20263 weeks ago-
DarwinX: Evolving Agent Harnesses Through Natural SelectionDarwinX evolves agent harnesses via population selection with frozen models, improving verified performance across benchmarks without benchmark-specific patches.Agents and evaluation · Jul 20268 weeks ago-
StateAct: Program State, before Pixels, for Long-Horizon Computer-Use AgentsStateAct improves computer-use agents by grounding actions, verification, and memory in direct program state rather than screenshots, boosting success rates while reducing cost.Agents and evaluation · Jul 20262 months ago-
Evidence-Backed Video Question AnsweringEvidence-Backed Video Question Answering requires models to provide answers with precise spatio-temporal segmentation evidence, revealing gaps between reasoning and visual grounding that are improved by large-scale instruction tuning.Multimodal and robotics · Jul 20262 months ago-
Learning from Language Feedback via Variational Policy DistillationVariational Policy Distillation enables reinforcement learning from language feedback by co-evolving teacher and student policies through variational expectation-maximization, overcoming limitations of passive distillation in complex reasoning tasks.Reasoning · May 20264 months ago-
The Illusion of Certainty: Decoupling Capability and Calibration in On-Policy DistillationOn-policy distillation suffers from miscalibration due to information mismatch between training and deployment contexts, which is addressed through a calibration-aware framework that improves both performance and confidence reliability.Training and scaling · Apr 20265 months ago-
Beyond the Assistant Turn: User Turn Generation as a Probe of Interaction Awareness in Language ModelsUser-turn generation serves as a probe to measure interaction awareness in large language models, revealing that this capability is distinct from task accuracy and can be influenced by training methods.Foundation models · Apr 20265 months ago-
GPA: Learning GUI Process Automation from DemonstrationsGUI Process Automation (GPA) offers robust, deterministic, and privacy-preserving vision-based robotic process automation with faster execution than current vision-language model approaches.Multimodal and robotics · Apr 20265 months ago-
Least-Loaded Expert Parallelism: Load Balancing An Imbalanced Mixture-of-ExpertsImbalanced expert routing in Mixture-of-Experts models leads to computational inefficiencies in expert parallelism, which are addressed by a dynamic rerouting algorithm that balances workload and reduces memory usage.Architectures · Jan 20268 months ago-
Agentic Confidence CalibrationAgentic confidence calibration addresses limitations of static calibration methods by introducing a trajectory-based diagnostic framework that improves reliability across diverse AI agent systems.Agents and evaluation · Jan 20268 months ago-
Agentic Uncertainty QuantificationA unified dual-process framework transforms verbalized uncertainty into active control signals for improved reasoning reliability in AI agents.Agents and evaluation · Jan 20268 months ago-
Future Optical Flow Prediction Improves Robot Control & Video GenerationA novel language-conditioned optical flow forecasting model combines Vision-Language Model and Diffusion architecture to predict future motion from noisy web-scale video data, demonstrating versatility in robotic manipulation and video generation tasks.Multimodal and robotics · Jan 20268 months ago-
LoCoBench-Agent: An Interactive Benchmark for LLM Agents in Long-Context Software EngineeringLoCoBench-Agent evaluates large language models as autonomous software development agents using interactive scenarios, specialized tools, and multi-turn conversations to assess their long-context performance, comprehension, and efficiency.Agents and evaluation · Nov 202510 months ago-
MMPersuade: A Dataset and Evaluation Framework for Multimodal PersuasionMMPersuade is a framework for studying multimodal persuasion in Large Vision-Language Models, revealing insights into their susceptibility and the effectiveness of various persuasive strategies across different contexts.Agents and evaluation · Oct 202511 months ago-
Enterprise Deep Research: Steerable Multi-Agent Deep Research for Enterprise AnalyticsEnterprise Deep Research (EDR) is a multi-agent system that automates report generation and real-time data analysis by integrating specialized agents and tools, outperforming existing agentic systems on open benchmarks.Agents and evaluation · Oct 202511 months ago-
Foundational Automatic Evaluators: Scaling Multi-Task Generative Evaluator Training for Reasoning-Centric DomainsFARE, a family of large-scale parameter evaluators, surpasses specialized RL-trained evaluators in both static benchmarks and real-world tasks through data-driven development and iterative rejection-sampling supervised finetuning.Reasoning · Oct 202511 months ago-
LiveResearchBench: A Live Benchmark for User-Centric Deep Research in the WildLiveResearchBench and DeepEval provide a comprehensive framework for evaluating deep research systems across various domains, focusing on real-time web search, synthesis, and citation-grounded long-form reports.Agents and evaluation · Oct 202511 months ago-
Hard2Verify: A Step-Level Verification Benchmark for Open-Ended Frontier MathHard2Verify, a human-annotated benchmark, evaluates step-level verifiers for LLM-based mathematical reasoning systems, highlighting the challenges and performance gaps between open-source and closed-source models.Agents and evaluation · Oct 202511 months ago-
Webscale-RL: Automated Data Pipeline for Scaling RL Data to Pretraining LevelsA scalable data engine converts large-scale pre-training documents into diverse question-answer pairs for reinforcement learning, significantly improving model performance and efficiency.Training and scaling · Oct 202511 months ago-
UNIDOC-BENCH: A Unified Benchmark for Document-Centric Multimodal RAGUniDoc-Bench is a large-scale benchmark for multimodal retrieval-augmented generation, evaluating systems across text, images, and their fusion in real-world document-centric scenarios.Agents and evaluation · Oct 202511 months ago-
CoDA: Coding LM via Diffusion AdaptationCoDA, a 1.7B-parameter diffusion coder, achieves competitive performance with smaller models through confidence-guided sampling and is released with open-source tools.Inference and efficiency · Sep 202512 months ago-
MCP-Universe: Benchmarking Large Language Models with Real-World Model Context Protocol ServersMCP-Universe is a comprehensive benchmark designed to evaluate large language models in realistic tasks through interaction with real-world MCP servers, addressing challenges like long-horizon reasoning and unfamiliar tool spaces.Multimodal and robotics · Aug 20251 year ago-
GTA1: GUI Test-time Scaling AgentA GUI Test-time Scaling Agent addresses task planning ambiguity and visual grounding accuracy in GUI interactions using reinforcement learning and test-time scaling.Agents and evaluation · Jul 20251 year ago-
Demystifying Domain-adaptive Post-training for Financial LLMsFINDAP, a systematic investigation, improves LLMs for finance through continual pretraining, instruction tuning, and preference alignment, achieving state-of-the-art performance using a novel preference data distillation method.Training and scaling · Jan 20251 year ago-
xLAM: A Family of Large Action Models to Empower AI Agent SystemsxLAM, a series of large action models, enhances agent performance by utilizing diverse datasets and achieving top results on benchmarks like the Berkeley Function-Calling Leaderboard.Agents and evaluation · Sep 20242 years ago-
CodeT5+: Open Code Large Language Models for Code Understanding and GenerationCodeT5+, a family of flexible encoder-decoder LLMs initialized with frozen LLMs and trained with a mixture of pretraining objectives, achieves state-of-the-art performance across various code-related tasks.Architectures · May 20233 years ago-
BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language ModelsBLIP-2 efficiently pre-trains vision-language models using off-the-shelf frozen encoders and decoders, achieving state-of-the-art performance with fewer parameters.Multimodal and robotics · Jan 20233 years ago-
BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and GenerationBLIP, a Vision-Language Pre-training framework, improves performance across both understanding and generation tasks by bootstrapping captions from noisy web data, achieving state-of-the-art results on image-text retrieval, image captioning, and VQA, and showing strong generalization to video-language tasks.Multimodal and robotics · Jan 20224 years ago-
CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and GenerationCodeT5, a unified encoder-decoder Transformer model, improves code understanding and generation by leveraging semantic information from identifiers and user comments.Architectures · Sep 20215 years ago-
About and links

Topics: Agents and evaluation 14, Multimodal and robotics 6, Architectures 4, Training and scaling 4, Reasoning 3, Foundation models 1, Inference and efficiency 1

Salesforce on Companies (companies.fru.dev)

Also onAcquisitions·Quarterly·Paydays·TechConf

New papers by email

Monday afternoons, only in weeks with new papers from the labs.

Double opt-in. Unsubscribe any time.