10 papers
EffGen: Enabling Small Language Models as Capable Autonomous Agents
Gaurav Srivastava, Aafiya Hussain, Chi Wang +2
Most existing language model agentic systems today are built and optimized for large language models (e.g., GPT, Claude, Gemini) via API calls; while powerful, this approach faces…
SLMJury: Can Small Language Models Judge as Well as Large Ones?
Anish Laddha, Nitesh Pradhan, Gaurav Srivastava
Large language models (LLMs) are widely used as judges for evaluating model outputs, but their high cost, latency, and opacity limit scalability. We introduce SLMJury, a framework…
Do LLMs Overthink Basic Math Reasoning? Benchmarking the Accuracy-Efficiency Tradeoff in Language Models
Gaurav Srivastava, Aafiya Hussain, Sriram Srinivasan +1
Large language models (LLMs) achieve impressive performance on complex mathematical benchmarks yet sometimes fail on basic math reasoning while generating unnecessarily verbose res…
BeyondBench: Contamination-Resistant Evaluation of Reasoning in Language Models
Gaurav Srivastava, Aafiya Hussain, Zhenyu Bi +5
Evaluating language models fairly is increasingly difficult as static benchmarks risk contamination by training data, obscuring whether models truly reason or recall. We introduce…
SoundBreak: A Systematic Study of Audio-Only Adversarial Attacks on Trimodal Models
Aafiya Hussain, Gaurav Srivastava, Alvi Ishmam +2
Multimodal foundation models that integrate audio, vision, and language achieve strong performance on reasoning and generation tasks, yet their robustness to adversarial manipulati…
Hindsight is 20/20: Building Agent Memory that Retains, Recalls, and Reflects
Chris Latimer, Nicoló Boschi, Andrew Neeser +4
Agent memory has been touted as a dimension of growth for LLM-based applications, enabling agents that can accumulate experience, adapt across sessions, and move beyond single-shot…